The Answer Is Transaction Costs

Why Bosses Don't Wear Bunny Slippers: TAITC

• Michael Munger

Share episode

Send us a text

Ever wondered why firms exist in a market-driven economy? This month's episode promises to unravel this question by diving deep into Ronald Coase's seminal 1937 paper, "The Nature of the Firm." Join me, Mike Munger, as I reflect on our first 16 months of podcasting and share the insights and wisdom that have shaped our journey. You'll gain a thorough understanding of how transaction costs influence economic behaviors and organizational structures, with fascinating examples from Richard Langlois' analysis of the American Midwest's agricultural sector before the railroad era.



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 will be the last of the weekly episodes over the summer because classes start at Duke today. For the next eight months during the school year I'll be doing longer form monthly interviews. People have asked how I choose people to interview. I invite people that I want to learn from and, honestly, I can learn from almost anyone. Today I'll take some stock of the first 16 months, answer some questions about the podcast and talk a little about the origins of the idea of transaction cost and why it's important. A new twedge book of the week, this week's new letter 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, but it's an imaginary system. There always are transaction costs when it is costly to transact, institutions matter and it is costly to transact. One question I get is about the intro music and the voices in the background about transaction costs. The first voice is Ronald Coase. The second voice is Douglas North. Both of them were giants, at least in my pantheon of scholars who worked on transaction costs. The music is Morning Mysteries. I don't know who the composer was. I got it under a non-exclusive license from artisticmediacom.

Speaker 1:

Title of the podcast. The Answer is Transaction Costs comes from a story that I told early on, but let me say it again briefly, because it involves Douglas North, my dissertation defense. I was nervous and was asked a question by one of my dissertation committee members, who was Douglas North. I didn't know the answer, so I did what economists always do when we don't know the answer, that is, I went to the board and started writing equations. The reason that this is effective is that's also what economists do when they think they do know the answer, and of course, that means that you can hide for a while. Fortunately, for me at least, doug saw through this and after about it seemed like forever, but it was probably no more than two minutes Doug raised his hand, interrupted me and said in a voice that you might use to address a well-loved but not terribly bright child Michael, the answer is transaction costs. The answer I was looking for is transaction costs, and I learned later, and other people also told me, that for Doug North it didn't matter what the question was. The answer is transaction costs, and as I went on to study political science and public choice, I concluded that's largely correct. So the title of the podcast is actually the answer to almost every important question. Doesn't matter what the question is. The answer is transaction cost. As an illustration, let me say that one of the first people who worked on transactions costs carefully Now, john R Commons, is the person who originated the discussion of transaction cost and economics, and I talked about that in an earlier podcast. One of the first who tried to be systematic about transaction cost was Ronald Harry Coase. Not many people know what the H stands for. It was Harry. In his 1937 paper Coase tried to answer a question and the question Coase tried to answer was this If markets are so great, why are there firms? So let me read a little bit from Coase's 1937 paper, the Nature of the Firm.

Speaker 1:

It is convenient if, in searching for a definition of a firm, we first consider the economic system as it is normally treated by the economist. Let's consider the description of the economic system given by Sir Arthur Salter. The normal economic system works itself For its current operation. It is under no central control. It needs no central survey Over the whole range of human activity and human need. It is under no central control. It needs no central survey Over the whole range of human activity and human need. Supply is adjusted to demand and production to consumption by a process that is automatic, elastic and responsive.

Speaker 1:

An economist thinks of the economic system as being coordinated by the price mechanism. Society becomes then not an organization but an organism. The economic system works itself. This does not mean that there's no planning by individuals. These exercise foresight and they choose between alternatives. It's necessarily so if there's going to be order in the system. So there is planning.

Speaker 1:

But this theory assumes that the direction of resources is dependent directly on the price mechanism. Indeed, it's often considered to be an objection to economic planning that planning merely tries to do what's already done by the price mechanism. Sir Arthur Salter's description, however, gives a very incomplete picture of our economic system. Within a firm, the description does not fit at all. For instance, in economic theory we find that the allocation of factors of production between different uses is determined by the price mechanism. The price of factor A becomes higher in location X than in location Y. As a result, a moves from Y to X until the difference between the prices in location X and location Y disappears, except insofar as that price compensates for other differential advantages. But in the real world we find there are many areas where this does not apply. If a workman moves from department Y to department X, that workman does not go because of a change in relative prices, but because he was ordered to do so by his superior.

Speaker 1:

Those who object to economic planning on the grounds that the problem is solved by price movements can be answered by pointing out that there is planning within our economic system which is quite different from the individual planning mentioned above and which is akin to what is normally called economic planning, end quote. So what's interesting about that and the way I try to illustrate it in class is that most of us work for firms. We do not daily respond to price movements. We have a boss. So it's unlikely that anybody gets home from work and says boy, prices were in a bad mood today. What happens instead was that they say my boss was in a bad mood today. Bosses give far more people more orders than prices do. Now at some point, operating in the background, the sort of meta price of profits and loss determine whether the things, the orders, the plans done by bosses in firms is correct. But most of the time these are little areas of planning within the larger market system. So why are there firms? The answer is transaction cost. It's hard to imagine me on a production line putting a bolt in the fender of a car and then auctioning it on eBay. Transporting it to the location where someone else is going to put the next part of the car on. The production line just goes down the line without regard to price, and people do what they are ordered to do. We may find out about whether those orders were good by profit and loss, as I said, but the idea that prices organize the entire system is just mistaken, and the reason for that, the answer to the question why is it that we have firms? Is transaction costs.

Speaker 1:

When I teach a political economy, I start with the neoclassical theory of consumption and talk about prices. I talk about consumption, the cost of labor, the cost of inputs, but most people work for a firm, a company, some large organization. They have a boss. So there are people that provide computer services, janitorial services, legal advice, all sorts of other things that are purchased by the firm. Now, the boss of a firm always has to make a make or buy decision. Are we going to make this internally or are we going to buy it from outside? So in such a setting you have to try to figure out if there's enough servers or computers. So your IT guy says no, no, we need to buy more computers. Now the manager decides what equipment to buy on the market probably not going to make computers themselves and pays real prices. But then probably the boss doesn't sell computer time to the staff, he gives it away to the staff. Is it worth it in terms of somehow increasing profits or raising the company's share values? The boss has to guess about that. Now maybe you can hire consultants, but that's not the same as consulting prices. Different bosses in different firms make different guesses. Some of the firms are profitable, some are not. No one knows specifically what choices lead to increased profits or to bankruptcy because prices are not operating directly.

Speaker 1:

Now suppose one day in one firm one manager, maybe on a whim, outsources the computer services or the janitorial services or the legal advice or the accounting, not to India or Ireland but just to another company across town or across the country. Boss signs a contract after taking bids from several companies that provide those services. Now the companies are forced by those scolding winds of market competition to provide excellent service at low cost. So the boss looked at the different prices in the bids offered in this competition and you learned something. You learned how much the service cost to provide and you learn how much money you can save by laying off the employees who used to provide that service in-house.

Speaker 1:

It's hard to fire employees, particularly since most employees are smart enough to work hard enough to get acceptable performance reviews. The boss also has a hard time motivating the in-house staff. Watching each employee's expensive and tiresome and it's hard to fire them. But it's easy to fire contracted employees because you just signed a new contract with a competitor. Why not let the market system do your motivation work? So let's suppose that our outsourcing boss sees the company's profits rise dramatically because he fired a bunch of in-house workers and contracted out. Because he fired a bunch of in-house workers and contracted out, stock price goes up 18% in six months and life for the boss is good.

Speaker 1:

So one day the boss has this crazy thought. He asks himself a question that's never occurred to him before why do we have any employees at all? Why have a building? Why not just sit at home wearing my jammies and my bunny slippers, having a nice cup of tea, and outsource everything? I can write contracts to buy parts inputs. I can pay workers to assemble the parts on a contract basis. I can use shipping companies to box and then transport the product. Well, now the boss is elated. He never really liked having all those people around anyway, always asking questions, looking for direction and expecting him to know the answer. He fires all the employees effective one month from now and he takes bids on all the design, parts, manufacture, assembly, shipping that people used to do and signs all those contracts On day 31,. After all, those wasteful, smelly employees are gone and the new contracting efficiency regime is in place.

Speaker 1:

The boss has a nice breakfast, pours his tea, puts on his bunny slippers, puts his feet up on the desk at home, checks his email. He notices that he has 1,239 new messages. Holy cow Turns on his cell phone. He sees he's missed 485 calls and texts. What the heck?

Speaker 1:

Turns out that coordinating all those contract employees, making all those different transactions work together in time and space, is really hard. So the manager shucks the bunny slippers, puts on his suit, hurries over to his primary contract supplier of inputs. What are you doing? We have a contract. Supplier says wait, who are you again? Boss gives his name frantically. None of the other contracts can be fulfilled. Until he gets these parts, the supplier checks. Oh yeah, right, I see the order. Sorry we're running behind. We'll have the stuff to you by the end of next week, maybe Monday the following week, within three days, of course, the boss is fired because the company is plunging into bankruptcy. Stock prices fall by 75% and a desperate effort is made to rehire most of the old employees back to their jobs.

Speaker 1:

So the point is that the boss can't just wear bunny slippers. An awful lot of what the boss does is not write contracts, because there's slippage and difficulty in overseeing contracts. It is cheaper to use direction and, let's be honest, central planning than it is to use the market system of buying everything on the spot market. Now Ronald Coase did make an interesting observation, and that was that the size of the firm is likely to vary with the amount of transaction costs. If transactions costs are very high, the firm will be large, because we'll want to internalize all those transactions. They'll be done by command and ownership of the upstream or downstream resource. If, however, transaction costs are low, I can rely on being able to purchase a lot of those inputs, or I can rely on the downstream retailer to provide the services that are contracted for, and the firm can be smaller. So over time we expect the size of the firm to depend on the marginal cost of organizing the last transaction, that is, the most expensive transaction in the market versus within the firm. Now there's an interesting example of the problem of transaction costs in a paper called the Secret Life of Mundane Transaction Costs, written by Richard Langlois, and he gives an example that I often use, but he says it so well that it is better if I use Langlois' and he gives an example that I often use, but he says it so well that it is better if I use Langlois' statement of it. So an example of transaction cost may help clarify the meaning.

Speaker 1:

In the American Midwest, before the coming of the railroad, as indeed through much of agricultural history, wheat was stored, shipped and traded by the sack. Each sack of wheat was the product of a specific, identifiable farmer, which meant that repeated trades could generate reputation effects that assured the quality of the grain in the market. At the same time, however, this mode of storage meant large transaction costs in the neoclassical or the old economic sense the transportation cost of shipping the sacks by wagon and river to St Louis or Chicago. Brokerage fees, insurance premia, the implicit cost of price volatility and poor information about market prices at the destination, and cost of the stevedores lugging bags from warehouse to barge to warehouse. Even the burlap bags themselves cost two to four cents apiece, holding shipping route constant. Most of these costs were arguably incurred on a per-sack basis as part of the process of exchanging title to the wheat. It was as if some, perhaps most, of the grain spilled out of the sack between farmer and miller. All of this changed with the coming of the railroad in the mid-19. All of this changed with the coming of the railroad in the mid-19th century. It quickly became economical to store and ship wheat in bulk using the newly invented mechanical grain elevator. This reduced neoclassical transaction costs dramatically, but it necessitated mixing together the grain of many different farmers. It destroyed the system of quality control that had relied on reputational effect from repeated transaction with identifiable farmers. To solve this problem, the Chicago Mercantile Exchange paid the costs of creating standardized categories for wheat and persuading farmers and buyers to adopt these standards. In addition, they needed to pay the cost of inspecting the weak for conformance to the standards, which they did by commissioning inspectors. The cost of this system are clearly the costs of establishing and maintaining property rights. The cost of establishing the standards, though, are not neoclassical frictions. They're instead incurred once and for all as a fixed cost and cease thereafter to enter into marginal calculations. But such fixed costs are, of course, related to frictional costs in that they are substitutes for them. By paying the one-time cost of standards, I can avoid paying stevedores by the sack End quote. Well, the point of the example and it's a really interesting one is that changes in technology may allow different institutional arrangements that reduce transaction costs and, in the case of grain, the commodification of literally the commodification. So the wheat becomes a commodity then, rather than a specific good. So wheat becomes a commodity then, rather than a specific good. A sack of wheat from Mr Smith is a particular boutique thing. I know it's good because it's from Mr Smith. If I buy a bushel of wheat on the Chicago Mercantile Exchange, it's a bushel of wheat, it is a commodity, it is homogeneous. However, it is important, then, to have some other way of solving the problem of trust. So there's a couple examples of how transaction costs might work. First, in the existence of the thing called the firm, which, if you were just to take neoclassical economics would sound like a strange thing. And then the effect of transaction cost on commodifying wheat. Whoa, that sound means it's time for the twedge. This is a joke about theory of value, or maybe it's more like a story. Engineers and scientists will never make as much money as reality TV stars. Now a rigorous mathematical proof explains why this is true. Postulate 1, knowledge equals power. Postulate 2, time equals money. Now, as every engineer knows, work divided by time is equal to power. Postulate two time equals money. Now, as every engineer knows, work divided by time is equal to power. Now, since knowledge is equal to power and time equals money, we can substitute and we get work divided by money equals knowledge. Now we just solve that equation for money Work divided by knowledge equals money. That is what we get. Is work divided by knowledge equals money. Now some of that makes sense. The more work you do, the more money you get. However, notice that knowledge is in the denominator. So as knowledge approaches zero, money approaches infinity, regardless of the work done. So the conclusion is the less you know, the more money you make. This week's letter. I may have missed this, but who are the speakers of the quotes that open every episode. I assume one is Douglas North. Well, yes, the first one is Ronald Coase and the second is Douglas North. I'm not sure the answer to my question is transaction cost, but the problem certainly is. Private high schools and colleges require candidates to submit a large application including standardized tests, teacher recommendation essays and transcripts. It's a lot of work to assemble, but the school can choose to review whatever they want, so that's not a cost to them. However, they interview every single student, which has a huge transaction cost. Many of these schools have acceptance rates at or below 10%, so a vast number of the interviews are essentially a waste. While there are some marginal candidates that may be worth interviewing in case it improves their application, surely most could safely be rejected. Pre-filtering by the rest of the application seems like it would have much lower transaction costs. Does the interview give that much signal about a candidate's merit? Is there some value to the interviewing itself, for example giving feedback? Perhaps it's a form of marketing, getting the student to the campus for the tour? Or maybe there just isn't time to pre-filter and then schedule the interviews. I'd appreciate any insight you have, because this just baffles me, thank you. Or maybe there just isn't time to pre-filter and then schedule the interviews. I'd appreciate any insight you have because this just baffles me. Thank you, rl. Well, rl, it is an interesting question why we spend so much time on some of these filters. Surely you're right. Though there's some people we're very likely to want to admit, for the most part there's a bunch of people obviously 90% in your example that are not going to be admitted. We just don't know who they are, and it does seem like it would be possible to interview only the marginal cases. I wonder if it is not a costly signal as a way to try to limit the number of applications although I may be wrong about that, because schools probably want to have as many applications as possible so they can say that they have a low acceptance rate. But it is true that someone who is sort of interested, but not really but they had really good test scores you might admit. If you make them go through the process of interviewing, then it's likely that they will attend your school if they're actually accepted. But I don't know the answer. Let me throw that open. Do other people know of instances where these colleges interview most or all, if they can, of the applicants? Duke does some interviews. I don't believe they interview all by any means, but that's just a part of the admission process. I don't know. It is certainly an interesting question in terms of transaction cost. Is it primarily trying to solve the problem of asymmetric information? That would be the traditional story. There's a bunch of applicants. Asymmetric information that would be the traditional story. There's a bunch of applicants. It is possible to fake or at least misrepresent one's quality in a written application. The theory might be that we can tell more and tell quickly, from an interview, but I don't know the answer. It's time for Book of the Week. This week's book is Lives of the Laureates 23 Nobel Economists. It was edited by Roger Spencer and David McPherson. It is, as the name suggests, a discussion with interviews with, and biographies of, 23 Nobel Prize winning economists. Some of them are very interesting, maybe a little bit surprising. Certainly a good summer read. It of them are very interesting, maybe a little bit surprising, certainly a good summer read. It is nice and episodic and I'm sure you'll enjoy it. Well, the next episode will be released on Tuesday, august 27th. That will again be the last of the weekly episodes. We will move to the long-form interview episodes. After that We'll have a new topic, some letters and, of course, a hilarious new twedge. That's next time on Tidy C.