Austrians vs. Chicago School vs. Samuelsonians: *Only* a Difference of Emphasis

A long, long time ago, I had the bold idea to work through three textbooks in economics from three different perspectives: Murray Rothbard's Man, Economy, and State, George Stigler's Theory of Price, and Mas-Colell, Whiston, and Green's Microeconomic Theory, or as I separate them: Austrian, Chicago, and Samuelsonian.

Boy, was I naïve then? I also set myself the goal of trying to translate these texts into short blog posts as a way to learn. While I gave up on the blog posts (to no-one's disappointment), I kept reading each book. Almost two years later I can report that I've finished working through each book. Worked through, not necessarily understood everything...

I chose this strange task, because I had always been fascinated by different approaches to economics. What does it mean to do economics? How have different people answered that problem? All these questions are generally things grad students aren't supposed to worry themselves with.

I couldn't help it.

While my perspective can never be completely balance to judge these three books, I worked hard to give each a friendly reading. As I'll explain, reading each with a friendly eye was extremely valuable. Maybe I even understand a thing or two from each book. I definitely understand more thoroughly why different economists believe different approaches are the right way to go. I'm more sympathetic to the disagreements in presentation and style. Honest people just have disagreements on what is important.

A Few General Thoughts

This post is about  my thoughts after studying each book. If you want to read this as a comment on different "schools" of thought, go ahead, but I'm focusing on the books. However, there is an inherent problem in reading too much into this, because I'm talking about textbooks, compared to pioneering research. Deal with that as you wish.

I came into this project most sympathetic to Rothbard's approach, but left sympathetic to Stigler's. At no point was I a fan of MWG, probably because grad school forced me to study MWG, while the others were simply "for fun." There is a dryness is MWG that is hard to handle, although it was more wordy and entertaining than a lot of the references I used for first year. What does that say about graduate school?

MWG is clearly the most advanced in terms of mathematics and is only workable for grad students. There is no value in studying it before you know the math. Stigler reads like an intermediate text these days, but funnier and clearer. Man, Economy, and State is basically an intro textbook, so it's accessible. Sorry, Rothbardians. It's true. That's not an insult.

Luckily for me though, Rothbard throws in comments that an intro student would glance over, but are food for thought for an economist. Anyone who tries to disregard his book because it seems simple is missing out on lots of insights. It was definitely worth going through for me, even after reading Mises's Human Action a few times. If you're someone who is sympathetic to Austrian economists and would read with an eye to learn from it, it is worth the read. If you're going to read it just to find holes, it's not worth the investment. Finding holes is too easy if the reader doesn't appreciate the Austrian approach.

Similarly, if you're going to read MWG and stop on page 116 because the term "social welfare function" appears without a sneer, fine. You're not going to learn. If you absolutely reject the use of mathematics in economics as a Mises Institute presenter said in a speech to students (48:00), the same thing will happen. You cut yourself off from a whole world of economists.

If you pick of Stigler and laugh at the fact that he doesn't define axioms to start from, you're missing out. Does everyone get my point? Read these books to learn from a grow, not criticize. That's my goal in general.

A Difference Of Emphasis

The rest of this post is where I am going to get in trouble...

While I went into this project expecting to see completely different approaches to economics, the books were remarkably similar. This might make sense when one remembers that each book is still within neoclassical economics, roughly Walrasian, Marshallian, and Austrian. Nothing is heterodox. However, I was surprised. I had learned economics on the internet where everyone spends their days criticizing how dumb the "other" side is.

No. Each method is valuable in its own way.

The distinction comes from each book's particular emphasis, what the authors think is important. The emphasis of MWG is on a completely closed, axiomatic system, defining every part of a model and solving for an equilibrium. For Stigler, the emphasis is on simple, intuitive models that can applied to many problems. For Rothbard, I see the emphasis on coordination of production through prices. Of course there is Rothbard's emphasis on his form of praxeology and being able logically derive economic theory. Yeah, yeah, yeah.

But this is a "peculiarity of presentation."

Speaking about various school of thought in 1933, Mises (p. 288) wrote that

these three schools of thought differ only in their mode of expressing the same fundamental idea and that they are divided more by their terminology and by peculiarities of presentation than by the substance of their teachings.

Just to see how many people I can make mad, this quote still applies to neoclassical schools of thought in 2015, as exemplified by Rothbard, MWG, and Stigler. Mises didn't believe this even by 1957. But I'll disagree with The Master.

Each book's main focus is what Hayek called the pure logic of choice. Given some axioms, derive implications about people's choices. Each books focuses on equilibrium, even though they may have different ways of dealing with disequilibrium. Each has their demand curves sloping downwards, although MWG and Stigler might allow for hypothetical "Giffen goods." Each talks about markets that work to coordinate everyone's actions.

Of course there are also some differences in the tools that people will get hung up on.

  • MWG's theorem/proof style is confusing to people who didn't study that type of math.
  • Rothbard's rejection of advanced mathematics, indifference curves, or cardinal utility might confuse many economists.
  • Stigler's hand-waving, trying to balance between rigor and applicability might turn both sides off.

Yes, if you want to read the three books only looking for the differences, there are plenty of faults in all books.

However, I consider these ancillary to the general topic. None of these should stop readers from being able to learn economics from each book. These are all still *micro* texts. They deal with individual, purposive action. The have buyers and sellers interacting in a market. They have models to capture this. They are working to understand how markets work. Again, all are within the same neoclassical tradition.

The authors change the presentation of the material, but I don't see the fundamental issues that arise from these distinctions. Certainly nothing that would make it impossible for all three groups of economists to learn from one another.

Austrians may disagree with omitting discussions about the structure of production. MWG fans might miss the formal game theory in the other two books. I'll argue that the more fruitful way to read this disagreements is not worry about it. MWG thinks game theory is one of the important topics; the other do not (and they are old). Rothbard spends 300 pages on production; the others do not. Each book has a set of priorities. Space is limited.

Now, I'm sure some people will say that I do not appreciate the importance of some of these differences. They may say that cardinal/ordinal utility is too important to just brush aside. Or math is too fundamental that anything without math isn't "real economics." Fine. Spend your days tearing down the other approach. I'm not sure that is fruitful.

If person A says that the important thing in economics is having a model derived from axioms of preferences, but person B says the important part is having an easily applicable model, how are we to decide? Certainly, it is not a matter of science. I have my thoughts on what are important things to include and what are not. However, I see that as closer to disagreements over favorite band than to substantive disagreements that would hinder learning from each other.

To all economists, this is a plea saying that "the other guy" is not as crazy as you might think. That is my main take-away from reading these books. Authors have thought through economics seriously, but happens to disagree with you. You are only hurting yourself if you don't learn from him.

Moving Forward

I am constantly looking for common ground for research. I want to use insights from all of these and will keep reading in each traditions, particularly their approach to the core theory through textbooks. It helps me understand the hidden assumptions in each framework and clarifies these ideas in my teaching. I'm working through Kirzner's Market Theory and the Price System, Becker's Economic Theory, and Krep's Microeconomic Foundations. Wish me luck! And tear me apart in the comments 🙂

16 thoughts on “Austrians vs. Chicago School vs. Samuelsonians: *Only* a Difference of Emphasis

  1. Interesting stuff! I think it's really cool you took the time to read through these and draw comparisons.

    My own experience in grad school (which ended in 2013) was more binary: We used M-W-G, Silberberg & Suen, Romer, etc. in our theory classes. We used Deaton and Muellbauer for demand and dozens of applied articles for production.

    At the suggestion of my major professor, I read several books by modern Austrians and a few dozen articles. Having done my undergrad in the finance department and learned economic intuition and some modeling (it wasn't financial engineering, it was more fundamental analysis and mgmt strategy), I didn't really see the value in the highly abstract world of Samuelsonian theory. I felt like I learned as much from my reading of Austrian works even though I didn't spend as much time on it.

    I think my finance background pushed me toward a more empirical, applied worldview. The Austrian perspective (and probably the Chicago perspective, though I haven't read much on it) fits into that worldview better. When I use mathematics in my work, I look at it as telling the reader what I did with actual data. I tend to shy away from spending a lot of time deriving theoretical results. I don't see that as necessary for coming up with a good testable hypothesis.

    One example I can think of off the top of my head is the idea that plant fertilizer is a risk-increasing input for farmers. The "problem" discussed in the literature is that farmers tend to over-apply that input relative to the "optimal" application. Why would farmers over-apply an input that was risk-increasing? When I heard this in my master's level production class, I was baffled. Fertilizer is not risk-increasing, it's variability-increasing. A Samuelsonian sees risk as variance because that's the way we deal with risk in our mathematical models. In reality, risk is about the frequency of downside deviations from a target or losses. Looking at the problem from this point of view, fertilizer isn't risk-increasing and "over-applying" it makes perfect sense because the tonnage of product that comes out of the field ultimately determines the farmer's performance.

    This is a case, I think, where laser-like focus on the math has really confused our analysis.

    IMO, the "causal-realist" aspect is what really sets the school apart. It puts the focus on empirical reality and cautions against focusing too hard on the "world of the model" for explanations/solutions relevant to reality.

    • I agree that the big distinction is not solely about the level of math used. Math can confuse people, just like words can. Although, I have a hard time imagining that the optimal amount of math for my own work is zero. I tend to eschew corner solutions.

      As far as Rothbard as "causal-realist," I'm not sure I ever understood the distinction. Rothbard wants to describe the real world. Many other economists do too. Yes, some economists work in a purely abstract world, but that's not a necessary consequence of doing non-Austrian work. I take Rothbard as arguing that models should relate to the real world. True. But someone following your Silberberg book, chapter 1, would have the same idea, but working with formal models.

    • Levi,

      I think that the optimal fertilizer example is spot on! I am often very skeptical of claims that academics know when growers are doing too much of anything. I've been wanting to dig more into the literature on "optimal" fertilizer applications (as it seems like some of the insights could play over into herbicides). Do you have any recommendations for where to start? Any papers that use the insights you just lay out?


      • Dallas,

        Paulson and Babcock talk about the issue in this paper:

        The second half of the last paragraph makes the point well: your model is going to misunderstand people's behavior when your assumptions are wrong. In this case, it's so bad that prediction is likely to be off.

        I have some ideas for a paper addressing the "challenge" presented in this paper. I'll send you an e-mail.

        • Awesome. Thanks! The paper looks interesting. Contrary to my earlier statement, I think this a theoretical issue that econ has not really resolved--understanding how people behave under uncertainty. There are lot of well known experimental and real-world anomalies where standard expected-utility-maximization doesn't fit the data very well.

          I just got your e-mail and will get back to you.

  2. Honestly, I wonder if the theory battles are over in micro. It seems to me like everyone is working with the same Micro tool kit now. The real difference is how you do the empirical work. At one pole you have structural econometrics and at the other pole you have reduced form econometrics.

    At least that's the impression one gets reading Angrist.

    • I'd probably agree with you point on theory. My impression is that the main theoretical differences (at least at the textbook level) are over what is worth studying and interesting. The main battles are now about what parts to emphasize, processes or equilibrium, etc, not whether businesses maximize profits.

      I don't dare venture to talk about applied work. My impression as an outsider is that substantive differences remain their.

    • I'm not sure if the battles are ever over, and I think it's pretty unhealthy for the profession if they are. I really like Boettke's mainline/mainstream distinction on issues like this.

      I would agree that the main battles are over empirical methods. Personally I'd like to see more diversity (and disagreement, if necessary) in the empirical methods. Expanding the use of the case study approach could be fruitful in some situations, but the probability of a journal publishing something like that in the better journals is practically nil.

      • Levi,

        Yah, I didn't mean to say the theory battles are totally over. I think they are just currently taking a back seat. It'll be interesting to see how empirical findings work their way back and alter theory. Behavioral is already having some influence on theory.

        Also, I agree that case studies don't get enough respect. However, I think they can occasionally find outlets. I think a lot of Peter Leeson's stuff could be described as case studies of specific historical institutions through the lens of rational choice theory. And he has certainly got a lot attn. My favorite example of this is Trading with Bandits but Ordeals is also a good paper in this vein.

        • Dallas,

          Good point on theory disputes taking a backseat. Bold statement: in a world where market process theory dominates, behavioral economics would not have developed because it wouldn't be necessary.

          Yes, you're right about Leeson's stuff. I read the Ordeals paper and enjoyed it very much.

Comments are closed.