It sounds to me like a big part of the problem is how one conceptualizes the purpose of an econ 101 course and how it is taught. If the role of the course is to introduce some simple models that will be useful to build on not a claim about causal structure this seems fine -- these models are to econ what the simple harmonic oscillator is to physics.
It seems like your concern is (reasonably) that people take these models as predictive of what one should see in real world situations. The models can be largely fine simple idealizations and because the number of firms in a market isn't really an independent variable tell us very little about what should be expected in terms of consumer surplus in those situations.
And this seems like a genuine tension. I tend to even agree with what you are saying in terms of what is really going on but I have trouble imagining how you make that work as a class. The whole problem is that if students haven't taken a class in which they manipulate and apply pretty simple models like this how do they understand the more advanced ideas that depend on apply more complex models?
---
This feels kinda analogous to the problem we have in mathematics where the kind of teaching that is going to work for someone who is interested in becoming a mathematican and thinking mathematically doesn't work for the students who don't like the subject and just want to be told how to solve their problems.
My view is that we need to more clearly break these courses into different tracks but the incentivizes just don't match. Every time you try to break things into a course that teaches math for those who are genuinely interested then taking that course becomes a signal for mathematical ability and the incentivizes favor lots of people taking it for that reason and that changes instructor incentives. Similarly, if you try to create two different tracks in econ with one that just basically tells them: here is what actually happens according to the research (even if you lack the modelling capacity to understand it) and another which teaches students how to use simple economic models the second will be seen as the 'serious' econ course with the same downstream issues.
There’s a question missing from this analysis: How do you define the market?
Standard Oil once had a near monopoly on the refining of kerosene, which was used for lighting. But it didn’t have a monopoly on kerosene substitutes such as coal gas and electricity.
Defining the market as kerosene is very different from defining it as lighting. A regulator who correctly understands Demsetz can still cause a lot of harm by defining the market too narrowly.
“It also naturally leads into questions of why Firm X is a price-taker while Firm Y is a price-searcher.”
Your questions brings to mind Ronald Coase.
Ronald Coase:
“Economists always go on and on about the price system and its magic. But if markets and prices are so great, why are there firms? But then, if firms are so great, why is there not just one big firm?
Coase’s answer is now standard in economics: Firms will expand or shrink until the cost of “making” equals the cost of “buying.”” (Michael Munger)
This statement is a core summary of Ronald Coase's 1937 landmark essay, "The Nature of the Firm," which explains why businesses exist in a market economy rather than everything being produced by independent contractors.
Coase argued that while the market is efficient, it is not free to use; it costs money, time, and effort to search for suppliers, negotiate contracts, and enforce agreements—collectively known as transaction costs.
Are you arguing that under monopoly, its possible for firms to be price-takers with no ability to influence the market price and in contrast, under perfect competition, its possible for firms to be price-setters with significant market power that allows them to charge high prices and earn large profits?
I don't see how the first is possible. The second is possible because firms collude and act like a cartel.
The argument for the existence of the second is if 1. the monopoly is not free to drop prices and 2. if the monopoly sets prices above the competitive equilibrium, a new firm will immediate enter, pushing prices to zero. The monopoly either tries raising prices, producing the challenge, but more likely does not bother, and so is a price taker competing so that the next plausible entrant does not enter, even though the next plausible entrant is not counted among the firms in the industry, as it has zero market share and otherwise exists only in the mind of capital allocators. Or B. one firm, having outcompeted the second to last firm in a brutal price war which kept the market at price taking, does not change it's behavior, and does not realize profits. Soon the situation above holds. The key word is contestable markets
I’d argue we need to go back to the fundamentals of teaching. What is the purpose of 101? Has it evolved over time as economic thinking has changed? I once said to a student, who was a very good maths-science student, “you might like the challenge of economics because in social sciences we can’t control the lab the way you might in the natural sciences.”
The student, very bright now credits me with ‘converting’ her to economics. We want to analyse decision-making, however, we are more excited about predicting using mathematical models. Let’s get the analysis right first. In turn the predictions will be better.
Increasingly, I think the core ideas of Alchian's 1950's paper on "Evolution..." would be helpful for intro students to understand the role of firms in a market economy. Other Alchian papers on search cost and information/transactions costs develop ideas that are essential to a fuller treatment of theory of the firm. Ultimately, understanding that in economics a "market" is the utilization of dispersed knowledge, and that property rights are essential to a market economy should be the core of Econ 101.
I’m not too familiar with IO but I have a hard time believing that empirical IO is out there uncovering the true causal fabric of the economic universe. Much more likely is that empirical IO today suffers from the same methodological flaws that Cartwright and many others have identified in empirical micro more generally.
With that in mind, I disagree with your point here. Do market structures not cause firm behavior? In my opinion, we have no idea what is the right way to think about these questions. But thinking in terms of market structure is simple and still communicates practically useful ideas.
A massive portion of economics taught in school does not accurately describe the real economy that actually exists. Econ 101 ideas mostly don’t fall on the good side of that ledger
Great post. I've been playing around with somewhat similar ideas in my head recently. I wonder if dropping the current version of the supply curve entirely makes sense and instead just making marginal cost the primitive object. Then you can tell a story about how competition can drive price down to marginal cost or market power can result in prices above that marginal cost. Basically just reverse the standard story. Instead of introducing supply and then later explaining how it relates to marginal cost, start with marginal cost and then explain how it relates to supply.
You can still get all of the same comparative statics by shifting the marginal cost curve instead of shifting a "supply curve" except now it also works even in cases with market power (just the magnitudes would be different).
Not sure if this fully makes sense. Would be curious to hear your thoughts.
It sounds to me like a big part of the problem is how one conceptualizes the purpose of an econ 101 course and how it is taught. If the role of the course is to introduce some simple models that will be useful to build on not a claim about causal structure this seems fine -- these models are to econ what the simple harmonic oscillator is to physics.
It seems like your concern is (reasonably) that people take these models as predictive of what one should see in real world situations. The models can be largely fine simple idealizations and because the number of firms in a market isn't really an independent variable tell us very little about what should be expected in terms of consumer surplus in those situations.
And this seems like a genuine tension. I tend to even agree with what you are saying in terms of what is really going on but I have trouble imagining how you make that work as a class. The whole problem is that if students haven't taken a class in which they manipulate and apply pretty simple models like this how do they understand the more advanced ideas that depend on apply more complex models?
---
This feels kinda analogous to the problem we have in mathematics where the kind of teaching that is going to work for someone who is interested in becoming a mathematican and thinking mathematically doesn't work for the students who don't like the subject and just want to be told how to solve their problems.
My view is that we need to more clearly break these courses into different tracks but the incentivizes just don't match. Every time you try to break things into a course that teaches math for those who are genuinely interested then taking that course becomes a signal for mathematical ability and the incentivizes favor lots of people taking it for that reason and that changes instructor incentives. Similarly, if you try to create two different tracks in econ with one that just basically tells them: here is what actually happens according to the research (even if you lack the modelling capacity to understand it) and another which teaches students how to use simple economic models the second will be seen as the 'serious' econ course with the same downstream issues.
I don't have a good solution.
There’s a question missing from this analysis: How do you define the market?
Standard Oil once had a near monopoly on the refining of kerosene, which was used for lighting. But it didn’t have a monopoly on kerosene substitutes such as coal gas and electricity.
Defining the market as kerosene is very different from defining it as lighting. A regulator who correctly understands Demsetz can still cause a lot of harm by defining the market too narrowly.
“It also naturally leads into questions of why Firm X is a price-taker while Firm Y is a price-searcher.”
Your questions brings to mind Ronald Coase.
Ronald Coase:
“Economists always go on and on about the price system and its magic. But if markets and prices are so great, why are there firms? But then, if firms are so great, why is there not just one big firm?
Coase’s answer is now standard in economics: Firms will expand or shrink until the cost of “making” equals the cost of “buying.”” (Michael Munger)
This statement is a core summary of Ronald Coase's 1937 landmark essay, "The Nature of the Firm," which explains why businesses exist in a market economy rather than everything being produced by independent contractors.
Coase argued that while the market is efficient, it is not free to use; it costs money, time, and effort to search for suppliers, negotiate contracts, and enforce agreements—collectively known as transaction costs.
Are you arguing that under monopoly, its possible for firms to be price-takers with no ability to influence the market price and in contrast, under perfect competition, its possible for firms to be price-setters with significant market power that allows them to charge high prices and earn large profits?
I don't see how the first is possible. The second is possible because firms collude and act like a cartel.
The argument for the existence of the second is if 1. the monopoly is not free to drop prices and 2. if the monopoly sets prices above the competitive equilibrium, a new firm will immediate enter, pushing prices to zero. The monopoly either tries raising prices, producing the challenge, but more likely does not bother, and so is a price taker competing so that the next plausible entrant does not enter, even though the next plausible entrant is not counted among the firms in the industry, as it has zero market share and otherwise exists only in the mind of capital allocators. Or B. one firm, having outcompeted the second to last firm in a brutal price war which kept the market at price taking, does not change it's behavior, and does not realize profits. Soon the situation above holds. The key word is contestable markets
I’d argue we need to go back to the fundamentals of teaching. What is the purpose of 101? Has it evolved over time as economic thinking has changed? I once said to a student, who was a very good maths-science student, “you might like the challenge of economics because in social sciences we can’t control the lab the way you might in the natural sciences.”
The student, very bright now credits me with ‘converting’ her to economics. We want to analyse decision-making, however, we are more excited about predicting using mathematical models. Let’s get the analysis right first. In turn the predictions will be better.
Increasingly, I think the core ideas of Alchian's 1950's paper on "Evolution..." would be helpful for intro students to understand the role of firms in a market economy. Other Alchian papers on search cost and information/transactions costs develop ideas that are essential to a fuller treatment of theory of the firm. Ultimately, understanding that in economics a "market" is the utilization of dispersed knowledge, and that property rights are essential to a market economy should be the core of Econ 101.
I’m not too familiar with IO but I have a hard time believing that empirical IO is out there uncovering the true causal fabric of the economic universe. Much more likely is that empirical IO today suffers from the same methodological flaws that Cartwright and many others have identified in empirical micro more generally.
With that in mind, I disagree with your point here. Do market structures not cause firm behavior? In my opinion, we have no idea what is the right way to think about these questions. But thinking in terms of market structure is simple and still communicates practically useful ideas.
I'm intrigued and scared to learn what is feminist economics
And physics 100 years. Wanna teach quantum or GR out of the gate?
Thanks, Brian. I'm teaching imperfect competition this week and this was a helpful reminder of how I frame things in class.
A massive portion of economics taught in school does not accurately describe the real economy that actually exists. Econ 101 ideas mostly don’t fall on the good side of that ledger
Great post. I've been playing around with somewhat similar ideas in my head recently. I wonder if dropping the current version of the supply curve entirely makes sense and instead just making marginal cost the primitive object. Then you can tell a story about how competition can drive price down to marginal cost or market power can result in prices above that marginal cost. Basically just reverse the standard story. Instead of introducing supply and then later explaining how it relates to marginal cost, start with marginal cost and then explain how it relates to supply.
You can still get all of the same comparative statics by shifting the marginal cost curve instead of shifting a "supply curve" except now it also works even in cases with market power (just the magnitudes would be different).
Not sure if this fully makes sense. Would be curious to hear your thoughts.