The Answer Is Transaction Costs

When Bats Attack: Understanding Insurance

• Michael Munger • Season 3 • Episode 10

Share episode

Send us a text

Mike Munger explores insurance economics through the lens of transaction costs and risk management, culminating in an amusing case study about "bat-in-mouth disease."

  • Insurance transfers risk from individuals to larger pools, reducing the expected variance of outcomes
  • The fair price of insurance equals expected value (probability × potential loss) plus transaction costs
  • Information asymmetry, subjective risk valuation, and strategic behavior complicate insurance markets
  • Insurance faces two major challenges: adverse selection (who buys insurance) and moral hazard (behavior changes after getting insurance)
  • Deductibles and co-pays help align incentives between insurers and insured
  • Insurance history dates back 5,000 years to ancient China, Mesopotamia, Greece, and Rome
  • The "bat-in-mouth disease" case study shows what happens when someone tries to purchase insurance after an incident
  • Transaction costs explain why dogs sometimes stop climbing stairs and why freezing credit cards--ie, transaction costs--might prevent impulse spending. The piano player in a brothel story, and its history.
  • The book o'da'month is Daniel Flynn, The Man Who Invented Conservatism. 


Bat in mouth story:  https://www.nbcnews.com/news/us-news/bat-flies-womans-mouth-arizona-costing-nearly-21000-medical-bills-rcna222463

Some background on insurance:

"Piano player in a brothel" story origins:

Daniel Flynn book:  The Man Who Invented Conservatism


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 this week insurance and a case study based on an actual example of bat-in-mouth disease. A new twedge. This week's letter book it a week and more. Straight out of Creedmoor. This is Tidy C.

Speaker 1:

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. Now the idea of insurance is very old.

Speaker 1:

Insurance is when I get someone else to accept the risk of the consequences of my actions or of revelations of different states of nature. The way that insurance works is that it's a financial contract that pulls risk across many individuals. So the way that I can make money by accepting risk is that I accept the risk of many people. If those risks are independent, then only a few of those people suffer a loss, but I collect revenues from all of them. However, each of those individuals is better off because they would rather pay a small amount and be insulated from the possible risks ex ante, although they may regret it ex post.

Speaker 1:

I have often said in a joke my wife doesn't find very funny when she says she's not sure we need some kind of insurance. I say, well, we have a life insurance policy on me. We've never used yet that's true, because I haven't died but we didn't know that at the time we got the life insurance policy. The core economic purpose of insurance is to transfer risk from an individual to a larger pool, which reduces the expected variance of individual outcomes. So the many who don't suffer any damage are a little bit worse off because they pay a fee. The one or two or seven individuals in the group who do suffer a loss have that loss mitigated by payment from the insurance fund, and the person who acts as the go-between or broker reduces the transaction cost of that exchange by organizing the money that comes in and paying the money out, provided it meets certain conditions. However, there are some big problems, and it won't surprise you to learn that they are largely transaction cost problems in insurance markets. One of these is information asymmetry. Consumers often don't fully understand what coverage includes or excludes and in fact it is in the interest of the insurance seller of the insurance product that we're paying for to disguise that. Now, that's in many ways up to the person who's buying the insurance and, to be fair, often it may be necessary to specify in ways that seem excessively precise and even cruel that this kind of coverage will be paid. This kind of coverage will not be paid and if you wanted the additional coverage you would have had to pay more. So saying after the fact oh, I thought that was covered. Well, you're obliged to look at the policy.

Speaker 1:

Second important issue is the subjective value of risk Insurance markets price how much it costs to lay off risk. People have to ask themselves how much they're willing to pay to have someone else accept that risk. So a cautious person might over-insure and a person who's a risk taker might under-insure. Now you might ask compared to what? And the fair price of insurance is the expected value. So if there's a one in 10 chance that I'm going to incur a loss of $1,000, and there were 10 of us we would expect each of us to pay $100 into this fund. One of the 10 of us on average, is going to suffer the loss and the $1,000 will go to them. Now each of us might have to pay one or two or 10 additional dollars in order to pay the insurance agent or broker for organizing this contract, but the fair value of insurance has to be at least the expected value. So I think a lot of people don't understand that when it comes to excessively risky contracts.

Speaker 1:

Suppose I want to live in a fire-prone or flood-prone or hurricane-prone area. I have a million-dollar house. There's a one-in-three chance that my house will be destroyed by a fire, earthquake or hurricane. What should the price of that insurance be? Well, each year there's a one-in-three chance that it's going to be completely wiped out and the insurance company is going to have to pay me a million dollars. That means that each year I'm going to have to pay at least $330,000 for that insurance policy. Now, of course I'm exaggerating the probability. There's no place that has a probability that's that high.

Speaker 1:

But a lot of consumers, I think, are very angry that they can't buy insurance. But remember, they expect the full value of their house to be replaced in the event of a loss. If the probability of the loss times the amount of the loss is a lot of money. That's what the insurance agent has to charge. There's no way around that. And they have to actually get enough more that it compensates them for taking the risk, because sometimes these probabilities are not independent.

Speaker 1:

A whole bunch of houses in one neighborhood, after all, are likely all to suffer if there's a fire, a hurricane or an earthquake, and so insurance companies can't have all of their insurance contracts written in the same market. They have to have them in different markets where the statistical probability of losses is independent. The problem is they're going to go bankrupt as soon as there's a law. That's actually fraud, and so it's very difficult, a very difficult problem, to work around the fact that the cost of insurance has to be at least the annual expected value. So the annual cost of an insurance contract has to be at least the annual expected value of a loss. And the way to take the expected value of the loss is the probability of the loss in that year times the value of a loss. And the way to take the expected value of the loss is the probability of the loss in that year times the value of the loss, and if that's a lot of money, insurance is going to be really expensive. There's no way around that.

Speaker 1:

Another transaction cost problem with insurance is already mentioned. In a way, insurers may deny claims or go bankrupt if they haven't set aside enough money to cover those losses, and so buyers are going to worry whether the promises are going to be honored and we try to use regulation to solve this problem, but it is a difficult, asymmetric information problem. One of the problems that people who write insurance contracts face is risk assessment how likely is the event that's being insured against and what's the expected payout? The skill that's needed to figure these problems out is actuarial, where you're multiplying probabilities and losses. So to stay solvent, insurance insurers have to price policies higher than the expected payout. It can't be less than and it can't be exactly equal to, because they have to cover their operating costs and a profit and their own need to cover against risks, because sometimes you're going to get events that are very low probability but all occur at the same time, are very low probability, but all occur at the same time.

Speaker 1:

Sellers also have the problem of fraud and strategic behavior. Sellers have to monitor the claims for dishonesty or manipulation. If I have a house and I'm going to go bankrupt and lose the house because I can't pay the mortgage, if I just set the house on fire, then I can use the insurance money to pay off the mortgage and so I have to worry. If I'm a seller, I have to worry that people are going to try to use the insurance as an asset to pay off other debts. So if you ask, how much should insurance cost?

Speaker 1:

Insurance is priced at the expected value of the payout adjusted for administrative costs. The expected value of the payout adjusted for administrative costs. A profit margin, which is the opportunity cost rate of return for the capital that's tied up in the business. The risk load, which is an uncertainty about the distribution of losses. And there are opportunities for cross-subsidization. Some groups are going to pay more to subsidize others, as in the case of unemployment insurance. Not all of us are equally likely to be unemployed, but if we all pay in the same amount for unemployment insurance, people who are very likely to be unemployed are paying too little. Those of us who are not likely to be unemployed are paying too much, but overall we each pay the same amount.

Speaker 1:

Many people object to having too much differentiation in insurance contracts between different risks, and so for that reason, for example, since men are much more likely to die relatively young and their average income may be higher, life insurance policies may be more expensive for men than for women. Most of the time we don't have much of a problem with that, but if medical costs are higher because women are more likely to give birth and have other variety of medical problems, then charging different medical insurance rates for men and women is seen as unfair. So the problem that we have is to what degree should we match the insurance costs with the expected risks, and one of the things that insurance companies try to do to compensate for this is to look for what are called pre-existing conditions. So if you want medical insurance and it turns out that you already have a disease, or even if you're likely genetically to have a particular disease, the insurance company would want to charge you a higher premium, but that seems unfair because I'm charging different people different premiums. It's not exactly unfair, though, because the amount is based on risk. So the question, the social question that comes down is what risks are we going to allow insurance companies to use to decide whether or not they're going to charge a particular price? So the right price, the fair price, is a function of actuarial fairness plus transaction cost, but many markets health, disaster, insurance are distorted by regulation or strategic behavior. Health insurance, in particular, is distorted by we insist that you cannot buy a health insurance policy unless certain things are covered, and that means that people who would prefer a cheaper insurance contract and not have those things covered or maybe have a higher deductible, are not allowed to do that because we restrict the contracts that insurers can offer, which suggests a question that I often ask my students and they almost always get it wrong. That I often ask my students and they almost always get it wrong, because it's hard to think in terms of compensating for risk.

Speaker 1:

My question is in health insurance, do you want a zero deductible health insurance policy if you have to pay for it? Do you want a zero deductible home insurance policy if you have to pay for it? Now, obviously, if it's free, if you don't have to pay for it, if it's paid for by taxpayers, of course, you prefer a zero deductible. Now, a deductible is the amount that you have to pay if there is a loss. So to what extent are you on the hook? And in medical terms, this is often a co-pay. So we'll have a small amount that the person has to pay and that is fixed, the amount then of the cost for the visit to the doctor. The insurance company pays everything else, but having a co-pay or having a deductible means that there is some reason for the person to take into account the likelihood of a claim. It's not free to me. I have to pay at least some small amount.

Speaker 1:

Well, for car insurance, do you want zero deductible? Well, let's think about it In order to get a zero deductible car insurance contract. That means that you have no risk. No matter what happens. If the car gets scratched, you don't have to pay anything. The insurance company will pay to repair it. Well, that means I have no particular reason to be very careful. I don't care if the car gets scratched Now. I have to take it to the body shop and then wait for a while, but I have no monetary cost. Suppose that instead I have a $200 deductible on body shop car repairs. Instead, I have a $200 deductible on body shop car repairs. Well, I'm more likely to be careful about small scratches in parking lots. I'm going to try to park it in places where it's not likely to be damaged. What if I have a $1,000 deductible? Well then, I'm likely to be very careful. So that sounds like I want a zero deductible so far. Ah, but wait, stop don't. A $1,000 deductible is going to be dramatically cheaper because the insurance company recognizes that, in effect, I'm self-insuring for the first $1,000. And that means that I can save more than $1,000 in expected value by accepting a $1,000 deductible, because I will behave differently Now.

Speaker 1:

This brings us to the two big problems that insurance companies face when it comes to the pool of people they're selling insurance to. Those two things are adverse selection and moral hazard. Adverse selection is when buyers know more about their own risk than insurers do. So if insurance is really expensive, who would buy health insurance? Well, only sick people would buy really expensive health insurance. But that means that the premiums are going to be driven up even more, driving out the few healthy people who were paying for expensive insurance. That leads to a death spiral in the market, because the insurance company is selling insurance policies only to people who have a higher expected value than the overall population, but they're the only ones who pay for it. So adverse selection is a sorting mechanism. So insurers combat this with medical exams, questionnaires, waiting periods, differentiated prices.

Speaker 1:

Second big problem is moral hazard. Moral hazard is when people change their behavior as a result of having insurance. So adverse selection is a change in the pool of people who are trying to get insurance. Moral hazard is a change in the behavior of the individual who has insurance, so a person with car insurance is going to be less careful. Somebody with theft insurance might be less careful about locking doors. So the way that insurers respond is with deductibles, co-payments, overall coverage limits. They're trying to maintain incentives for prudence, because the individuals, after all, know most about the problems that they're facing Now. Both of those issues stem from asymmetric information. Either the buyer knows more there's a problem of adverse selection or there's a change in behavior after the contract, which is moral hazard.

Speaker 1:

Well, are there problems with making insurance purchase mandatory? That is, everyone has to get insurance, and for automobile insurance we make it mandatory so that everyone who is driving a car can at least pay for the damages that they might do to others. So we're required to get liability insurance, and if you have a new car and you have a large car payment on that car, which means that the bank actually still mostly owns the car the bank might require that you have collision insurance, because then, if there's a collision, the bank can collect on the value of the car. And so there might be reasons why we want to make insurance mandatory. For health insurance, we might want to make insurance mandatory because of the problem of what we do with someone who doesn't have insurance. After all, if someone doesn't have insurance, we usually say that hospitals and emergency rooms are not allowed to turn them away, although something like that does sometimes happen if you don't have insurance and can't pay.

Speaker 1:

But that means that someone who takes the risk of having a catastrophic accident or getting very sick for something that requires large, expensive medical treatments over a long period of time is not going to pay for that potential risk. And they expect then, if they become injured or sick, for everyone else to pay for them. Or they just don't think about it, but they become injured or sick, for everyone else to pay for them, or they just don't think about it, but they're trying to free ride. So having mandatory insurance protects against free riding and aligns with an idea of social solidarity and risk pooling. Everyone has to pay into it and, as a result, everyone expects to benefit from it, and that's the way that large-scale systems that are called single payer, where everyone pays into it. And then there is social medical insurance. Everyone has, in effect, a social medical insurance policy.

Speaker 1:

But there are pretty good arguments against mandatory insurance, because forcing the purchase of insurance particularly in the US system, where you were going to be obliged to purchase insurance and that insurance had provisions that you might not have wanted to pay for. It's a way of avoiding the problem of adverse selection, but it does seem as if it violates personal autonomy. Mandates often involve implicit transfers, because low-risk individuals subsidize the high-risk ones. That can seem unjust if it's not clear and mandatory. Insurance laws may be lobbied for by insurance, inflating prices and limiting competition, because if you can get in the position of being the only company that offers this insurance and then make it mandatory, you obviously have a state-created monopoly.

Speaker 1:

Well, ideas of insurance are actually quite old. I tried to think of what some of the examples, the oldest examples of insurance, were. Examples the oldest examples of insurance were In ancient Mesopotamia, at something like 2000 BCE, so 4,000 years ago. The Code of Hammurabi mentions provisions for what were called bottomry contracts, that's B-O-T-T-O-M-R-Y bottomry contracts. So if a merchant's ship sank, the loans that had been used to fund that cargo had to be forgiven, which means that the risk of getting loans for a cargo are distributed between the ship owner and the people who had loaned the money in order to buy the cargo. Now that means, of course, that I'm going to require more interest or more payment in order to get a loan, but it does help distribute the risk.

Speaker 1:

A thousand years before that, in China, traders would spread their goods across multiple ships to reduce the risk of total loss, which is a rudimentary form of diversification and risk sharing. So of course I would like to have all of my stuff in a single ship, because it reduces the transaction cost of loading and unloading, but to the extent that different ships have an independent probability of sinking, running aground, being taken by pirates. If I split my shipment across multiple ships, even though the shipping part is more expensive, there's a kind of risk sharing and diversification of the probability of failing. So that's 5,000 years ago. Ancient Greek and Roman guilds had burial societies that would pool the resources so that members' funerals could be paid for and there might be a small pension or at least payment for the widow or the family of the member of the guild died. Their family would get some small compensation and this was pooled together from all. Everyone in the guild would pay into it and if someone died or were injured, they would take out a small amount of that in compensation. Muslim communities had cooperative risk-sharing structures that were collective organizations. They were trying to get around the Islamic prohibitions on gambling and interest.

Speaker 1:

And in the 17th century marine insurance Lloyd's of London, it was an impediment to the increase in the amount of shipping, especially with colonial shipping. And Lloyd's began as a coffeehouse where shippers and financiers met and what they were looking to do was to pool the collective risk and then if everyone paid in a small amount, only those who had losses would have it paid out. And there's a transaction cost problem, obviously. But if you can solve that, it means that the worst risks are substantially mitigated and the profits from having a lot of this kind of trade were large enough that they were able to begin to develop insurance contract. And so, of course, lloyd's of London is perhaps the oldest large-scale marine shipping insurance company in the world.

Speaker 1:

Now, those early institutions all show that the need for insurance, that having some ability to lay off loss and uncertainty are going to arise whenever trade, travel or community life made unpredictable losses, potentially catastrophic. You would like to have a way to reduce the amount of risk that you face and if you can get a group of people where the probability of loss is independent across the group, you can have mutually beneficial gains from trade, subject to the transaction cost of solving those contracting problems. So, overall, insurance is a really vital institution for managing uncertainty, but it faces some deep structural challenge. There's the pricing of risk. You need markets you need thick markets to price risk fairly, while dealing also with hidden information, and people often have the incentive to make sure information is hidden.

Speaker 1:

There's the behavioral problem of preventing insured parties from becoming reckless since they no longer bear the full cost of the risk of their behavior. There's the moral problem of balancing liberty and social welfare in mandate debates. Are we going to require people to participate in these essentially voluntary contracts? And there's the historical problem. It's interesting to look at the evolution from informal risk sharing to complex modern markets. That may be very difficult for people to understand because of the very large scale at which insurance operates. So the problem of insurance is an interesting ongoing one and it's one where the problem of insurance is an interesting ongoing one and it's one where the problem of transactions cost is pretty clear.

Speaker 1:

And that leads me to the example of what I had promised of bat in mouth disease. So apparently, a year ago, in August of 2024, there was a woman named Erica Kahn from Massachusetts, and she lost her job as a biomedical engineer and traveled to Glen Canyon National Recreation Area and was vacationing, taking some pictures, doing some hiking, and was there with her father, rich Kahn, and well, an awful thing happened. She was snapping some pictures of the night sky at Glen Canyon and a bat flitting around as bats do approached her and somehow got caught in between her camera and her face and it appears from this story that part of the bat got into her mouth. It is not clear to me. So she's holding her camera. The bat comes in between her face and the camera which she's holding to her face. The bat is flapping crazily, the wing of the bat goes into her mouth and from the picture, from the description, it sounds like part of the bat's wing came off in her mouth. Erica said it was kind of dark out. We were on a cliff, I was looking down and at my camera I didn't see the bat coming. It got tangled between my face and the camera. It was probably just a few seconds, but it felt like a lot longer.

Speaker 1:

So she was worried about rabies and there are very highly effective drugs for rabies but it can be fatal if it's not treated before the symptoms are felt, because you're not sure if you have rabies yet. So you have to go for the treatment, and the treatment is very expensive. Now she lost her job. She had declined to pay her former employee's insurance through COBRA, which is the Federal Continuation of Health Coverage Law, that is, you are required, the employer was required to allow her to continue to have insurance, but it would have cost her $650 a month, and she decided. According to the story, this unemployed Massachusetts woman figured she could roll the dice as a healthy woman in her early 30s or, at worst, could hastily buy private health insurance in a pinch end. Quote from the story Well, after the bat got into her mouth, khan went online, bought a policy and then went to get rabies treatments in Arizona, colorado and Massachusetts, believing that the insurance company would then pay for it.

Speaker 1:

So the bills come in and the total bills were nearly $21,000. And her policy had a 30-day waiting period before she could receive treatments that are covered by the plan. So remember the sequence of events here is she lost her job, she decided not to pay for the insurance that it was her right to receive, provided she paid for it. Bat flies between her face and her camera. Part of the bat is in her mouth. She goes and buys insurance and then goes and gets treatment.

Speaker 1:

Now, erica khan says I felt so powerless against these companies. It should be a human right to have life-saving care covered. In most other countries, like in europe, you just go to the hospital, you get your rabies vaccines and you pay nothing. Well, okay, but you're not ma'am and you had the chance to have insurance and you decided you didn't want it. So now what you want to do is go buy an insurance policy for, let's say, $500 a month. I'm not sure what it was. It was probably pretty expensive because individual rather than group, private medical insurance is pretty expensive. Let's suppose it was $500 per month. So she paid $500, and then she submitted her bill for $21,000, and the insurance company refused to pay because there's a 30-day waiting period. And she actually this event, this injury, had happened before she purchased insurance at all. This injury had happened before she purchased insurance at all. Now it would be a little more upsetting if she had bought insurance and the following day the bat had flown into her mouth and they said nope, we're not paying, there's a 30-day waiting period. But in this case the only reason she got insurance was that a bat flew into her mouth and she didn't have insurance. So her theory is she should be able to pay for one month of insurance after the fact and have all of her medical bills covered.

Speaker 1:

Now Khan is employed again. She has health insurance, but the gap meant that she is burdened by those bills. She said she should have secured private health insurance as soon as she was laid off. She should have done COBRA, even though it's very expensive. Well, is it very expensive? In fact, this is an illustration of how it probably is. Just something like the expected value. So the cost of private health insurance has to cover the expected cost of health care, and $650 a month is quite a bit for people that are not otherwise in a group.

Speaker 1:

You have a problem of adverse selection. What is it that has led this person not to have health insurance? So this is an illustration of several of the problems that we've talked about, and I also have to admit that I just like saying bat-in-mouth disease, and so that's why I brought up that example. Whoa, that sound means it's time for the twedge. I have two jokes this week, since we've been missing some twedges. First was suggested by my good friend, tony Gill. I had put on Facebook a picture of our dog, minga, who was going upstairs but became tired and so took a nap on the landing. Tony Gill suggested why did the dog not go upstairs? Well, the answer is transaction cost, and this probably happens pretty often. You could do a cost benefit analysis. I'm downstairs, I'd rather be upstairs, but you have to go upstairs, which means that the transaction cost of moving from downstairs to upstairs is enough that I may not act in a way that seems to benefit me. Now we might be able to use this to our benefit.

Speaker 1:

Some people have suggested, for those who have trouble with overspending, you don't need to get rid of your credit cards. What you need to do is freeze all of your credit cards in a large block of ice, and so you want to use the credit card. But in order to use it, you have to let the ice melt, and it's a big block of ice in your freezer. It takes an hour, and that additional transaction cost may be enough to prevent you from doing this action, which, in advance, you don't want to do, but you're worried about having at the spur of the moment. Well, the second twedge goes like this. This is from a letter from R. R suggests that I'm in my early 50s. I think you need to live through history a bit before you can appreciate this joke. So here is ours version.

Speaker 1:

Grade school teacher was asking students what their parents did for a living. Tim, you go first. What does your mother do? Tim stood up and proudly said she is a physician. Well, that's wonderful. How about you, amy? Amy stood up shyly, scuffed her feet and said my father is a mailman. He delivers the mail. Thank you, amy. What does your parent do, billy? Billy proudly stood up and announced my daddy plays piano in a brothel. The teacher was aghast. That evening she went to Billy's house and rang the bell. Billy's father answered the door. Teacher explained what his son had said and asked for an explanation. Billy's dad said oh yeah, well, I'm an economics PhD. How can I explain a thing like that to a seven-year-old? So I just tell him something else that's not as embarrassing. Well, thank you are.

Speaker 1:

That was from a letter, obviously, and it also is a pretty good twedge. What I like about that, as both a twedge and a letter, is that it raises a joke that I've heard before and I wondered about the history of it. So I went and looked. It appears that Harry Truman said of himself my choice early in life was either to be a piano player in a whorehouse or a politician, and to tell the truth, there's hardly any difference. And that joke. Versions of that joke also were told during the Great Depression. I actually found a blog post about it by Barry Poppock on his blog, who gives quite a bit of the history of the joke, but the earliest one that he was able to find was about bankers, and the problem was that being a banker in 1933, 1934 was kind of embarrassing. 1933-1934 was kind of embarrassing. So the way that the joke went and this was in a newspaper, don't tell my mother I'm a banker, she thinks I play piano in a and different versions of it bordello, brothel, cat house, house of joy, whorehouse is a joke about how the little people thought about bankers. So I'll put up the link to that blog post in the show notes.

Speaker 1:

Well, that brings me to the book of the week. This week I want to suggest a really excellent book by Daniel Flynn, the man who Invented Conservatism the Unlikely Life of Frank Meyer. It's about the origins of fusionism, which is a topic that I think is very interesting and it is a topical topic right now. It was published by Encounter Books in 2025. Well, that's it for this week. We'll have next week another episode of the Answer is Transaction Cost. We'll have another hilarious twedge, another letter, another book of the week and more Thanks for listening. Talk to you next week.