On intro micro first, and splitting micro and macro

I thought Noah Smith's post on why we should teach intro micro before intro macro was very good. I agree with Noah: even though macro is much more glamorous, micro is just as useful and important, and we are more confident that micro works. (Read his post for the full story.)

I want to add a couple of points in support of Noah, then talk about what I think is a bigger issue: the splitting of micro and macro into two separate half courses. I'm against splittism, but it's getting harder and harder to fight off the separatist forces.

Unlike Noah, who's just a young guy (judging from his photo), I speak from the vast authority of having taught intro (both micro and macro) about 30 times. So listen up, you kids.

Here are my couple of (related) points in support of Noah:

1. One of the most important things we teach in macro is the fallacy of composition. What is true of each of the parts isn't necessarily true of the whole. Just because the supply curves for apples, bananas, and carrots are all non-vertical does not mean the aggregate supply curve is non-vertical. The reasons the aggregate demand curve slopes down (if it does) are different from the reasons the demand curves for apples, bananas, and carrots slope down (except in a small open economy with a fixed exchange rate where it's basically the same reason). We can't just add up the supply and demand curves, because the things we are holding constant when we draw the curves for apples (like the price of bananas) are changing when we move along the curves for bananas. Each of those curves holds income constant, but income is changing as we move along the AD and AS curves.

If they didn't already understand micro, they wouldn't be able to understand what it means to add up micro to get macro. We wouldn't be able to properly teach the fallacy of composition.

2. The biggest of all macro questions is this: if you add up all the micro bits properly, so you avoid the fallacy of composition, why the hell does it seem not to work at the macro level? Or only work in the long run? (Or maybe it does work, if you are an RBC theorist who is determined to make it work?) It would be like if Newtonian mechanics worked OK for each individual car, but failed to work OK when you put a lot of cars together on a highway system.

If they didn't already understand micro, they wouldn't be able to understand that we have an empirical problem when we correctly (avoiding the fallacy of composition) add up micro to get macro. We wouldn't be able to properly teach the big macro problem of the trade cycle.

Now for the separation of intro micro and macro into two half courses.

At Carleton we have always taught intro as one full course. We do micro in the Fall term and macro in the Winter term. Usually, but not always, with the same instructor teaching both halves. But we face a lot of pressure from administrators, and students, to split it into two half courses, like most other Canadian universities. I can understand that pressure.

Maybe a transfer student has already taken micro at another university, and wants to complete just the macro half at Carleton, without repeating the whole course. Or vice versa. (We finally broke down and let them do this, because it was too embarrassing sending them down the canal to take the other half at University of Ottawa.)

Maybe a student is doing fine in the first half, but something goes wrong in the second half (maybe something totally unrelated to academic performance). Why can't we at least give credit for intro micro?

Maybe one half the course doesn't work with the student's timetable this year.

And so on. Everything is just so much more convenient all around if courses are broken up into easily-digestible bite-sized chunks. (The main data system, from the US, couldn't even grasp the concept of a two-term course, and had to be specially programmed to make it work.)

But I don't like the idea of splitting micro and macro into two half courses. It's just wrong. And it's wrong both ways, so simply making micro a pre-requisite for macro isn't the solution. Just as students can't understand macro without first understanding micro, they can't really appreciate micro without seeing where it fits into (or doesn't fit into) the big macro picture. (In that sense, I accept that the macro-first people do have a point.) I don't want to let any student out into the world until they have some idea of both micro and macro. They just do not have an introduction to either micro or macro unless they have seen the other side of the picture too, so they can understand how inadequate it is to see only one side of the picture. A little learning is, sometimes, a dangerous thing.

But I know I am fighting a lonely rearguard action to slow down the forces of progress and commonsense and what's best for students. They know I will retire soon, and will then do the sensible thing.

(P.S. I think it's correct to say that most teachers of intro, and most authors of intro textbooks, are more macroeconomists than microeconomists. Macroeconomists are more comfortable teaching basic micro than microeconomists are teaching basic macro. I think that tells us something important, though I'm not sure what.)

Comments

[justin]

"they can't really appreciate micro without seeing where it fits into (or doesn't fit into) the big macro picture"

One thing to consider here is people in many other disciplines have a need for micro, and this need would be just the same if macro didn't exist as a field. For instance, those studying marketing or industrial engineering should be able to have vast appreciations for micro without seeing how it fits in to the macro picture.

[Nick Rowe]

justin: yep. That's one of those commonsense what's best for students arguments that I am fighting my lonely rearguard action to resist.

[Peter K.]

As an amateur who has been following macro debates since the early 90s - so I guess I'm about Noah's age - I see it in the following way. To use your example of Newtonian physics, micro is like Quantum Mechanics and macro is like Newtonian physics, in that micro is a subset of the less accurate, more general macro. However, if used properly, Newtonian physics works. They employed Newtonian physics to put a man on the moon. They didn't need Quantum Mechanics which of course does a better job of describing reality.

As Konczal and Krugman point out, many economists got the macro issues around the financial crisis and ensuing recession and recovery wrong. Many students will wonder why this is the case. Perhaps part of the problem was the textbooks.

[Jacques René Giguère]

Peter K; fun thing is I view micro and macro in reverse to your example. Newtonian physics is simple and straightforward. It's at the quantum level that things get weird. You can't add speeds. Time and space contract. A particle can be a wave and vive-versa, in fact both at the same time, sometimes. Your wave function means you are everywhere at the same time so an electron can jump a potential barrier hogher than its own energy. Nah. It didn't jump, it was already there via its wave function. Einstein-Podolsky-Rosen mean that particle are entangled... Quantum physicist can teach Newton anytime. The reverse isn't. Same as for macro guy teaching micro in their sleep while Chicago guys just can't understand macro (they mico in their heads, whatever they think.)

[Jacques René Giguère]

Nick: yes. Fallacy of composition. But also "money". And "time". Money being an intertemporal mean of purchasing power transfer via its memory function.
And the balance of payments. Nobody in a pure micro world would rejoice in a current account surplus. But in a macro world, people can and do because they can make the mistake that holding money instead of goods is better.
I am basically a micro guy but veered macro in the last few years because things are funnier there.
But I already came in favor of splitting money into /kraputnics" and "gloupnology", if you recall...

[Evan]

Rather than having your intro econ course last for a year, why not just double up on intensity? It seems to solve some of the issues Nick raised.

To clarify, at my uni a standard single semester course is worth 3 credit points with 3 hours of lectures a week. Couldn't you organize intro econ as a 6 credit, 6 hour a week course, with the first half of semester spent on micro and the second half on macro? Or, alternatively, run the micro and macro alongside each other - 3 hours a week each (this would take some planning, but is possibly the best of both worlds)?

Where I did my undergrad a standard course was worth 10 credit points. In econ and the business school, 3 hours of lectures a week got you the 10 credit points. But in science and engineering, it took 6 hours of lectures a week to get the 10 credit points. First year math courses were split into calculus and algebra streams, each with as much work as a single econ course but for only half the credit!

[Jacques René Giguère]

At Laval inh the '70's, we did what Evan suggest. Intro ran side-by-side.

[Matthew]

I can see both sides. As it happens, I'm teaching Macro right now to a class of students who haven't taken micro. Yes, I do have to talk about things that would otherwise have been in micro, like preferences and introducing supply and demand curves. But as you point out, the fallacy of composition means that these things aren't really all that clearly related to their micro-analogues anyway.

I think there is something to be said about the fact that microeconomics has broadly been much more successful than macro as a science. But, the micro that was successful isn't the micro we teach in principles. My experience is that principles micro students always leave the course thinking either that the free market is a magical fairy land that never ever has any failures whatsoever, or else thinking that free markets are always massive failures and that if you want it done right, you gotta get the government to do it. Neither view is remotely close to the truth.

By contrast, principles macro presents a much less stylized, much more data-driven view of economics, even if macro the research field doesn't quite live up to that promise.

[Jacques René Giguère]

Micro has reached Keynes dream: boring as dentists.
Though they sometimes forget their basics as if they were battling mocros.
Anyone writing "the economics of deregulation" not thinking about the effects of an empty Samuelson core?

[Stephen Gordon]

At Laval in the '70's, we did what Evan suggest. Intro ran side-by-side.

That's still the way it's taught at Laval. Principes micro and Principes macro are two separate half courses, and students are expected to take both in their first term in the program.

[Nick Rowe]

We have one double-speed section of intro, in the Winter Term. And almost all courses in the Summer run at double-speed too. (That means 6 hours of classes a week instead of the regular 3.) I taught at quadruple-speed in Havana, for logistical reasons. It has its downside, since there is less time to reflect, or catch up, or turn stuff around.

The side-by-side micro+macro is interesting. I had never thought about that. They are thinking about one while they take the other, which is good. But in some ways it is the worst of both worlds, since they have done neither when they take the other. Dunno.

[Donald A. Coffin]

Justin--I have taught economics to MBA students who have never had an econ course (it's somewhat surprising how many of MBA students that is). I would argue that marketing students need to know both micro and macro in order to understand what works in marketing and what does not, to understand how economy-wide changes affect individual product markets (e.g., why some sectors of an economy are more strongly affected either by slumps or booms--income elasticity of demand being a key concept here, along with income changes). The relevance of macro (and micro) for finance seems to me immediately apparent...

[marris]

Nick, can you point to a resource which answers "the biggest of all macro questions" ? I've looked at some macro texts, but they either ignore or dance around the answer. Is the answer "money is different frmo all goods"?

[Lord]

It is always important when teaching models to also teach their limits, to know when and where they apply and when and where they may not. Micro as classical physics is taught first but its limits are also taught, large masses and relativistic speeds, small masses and quantum forces, even though macro. That it is incomplete and only approximate along with some of its problems should be made evident from the start.

[Peter Dorman]

Last year I tried an experiment in a two-quarter "program" (what we call courses at Evergreen -- they're full-time and interdisciplinary). I did the first half of micro and the first half of macro in the first quarter and the second half of each in the second. It seemed to work pretty well. Students who left after one quarter and passed the exams got a generalized intro to econ credit, like they would in a one-quarter "economic problems" course. It helped (of course) that I could use chapters from my own textbooks, since each has a sort of once-through-quickly-followed-by-returning-in-greater-depth format. (This is especially true of the macro, since I wanted to get students into data and the econ blogosphere as quickly as possible.)

[anon]

ISTM that the answer to this question would also depend in no small part on what your theoretical approach to macro is. Maybe it would be best to teach growth theory and the modern, monetary-disequilibrium approach to macro as part of the first-semester micro course, then have a second-semester course focused on the system of national accounts (and economic data, more generally) plus assorted macro topics.

[Frances Woolley]

Nick, I think the discipline of economics is, in a sense, much bigger than it was back in the day. While you've been teaching intro econ, there have been thousands of economists working on stretching the frontiers of the discipline.

It's got to the point when the discipline of econ is so big that it's not possible for any one person to understand more than a part of it. One could argue (I don't know if this is correct, but I'll argue it for the sake of argument) that micro and macro are on their way to becoming separate disciplines. This is why there is it is becoming increasingly difficult to sell the intellectual argument against dividing the two halves of intro - people really do see the two halves as distinctly different entities.

[linda]

I haven't taught intro (either side) for years (decades?), but do teach intermediate micro. We've interviewed for teaching professor positions in the past while, and frequently have the candidates give a "typical" intermediate macro lecture as their job presentation - since macro is where we have a shortage of teachers. I've found the lectures incredibly confusing, and not because of the presenters: I just don't understand the content: it's vague. To be fair, when I teach general equilibrium in micro (the basic 2x2x2 model), I point out to students that it can get very complicated, and that is one reason why macro is so hard.

[marcus nunes]

Nick. I remember telling students many years ago that just as there were microfoundations of macro, there were macro foundations of micro.

[The Keystone Garter]

Macro is basically micro but with only a few variables. Employment, central bank interest rates, CPI inflation....I think it will shortly be obsolete without including tech advances/regressions (tech advances have been considered missing in macro by some for many decades) and efficacy of Crowns/civil-service. Education has been estimated to be 80% of human capital and employment only covers it a bit.
Macro is useful because we put very well-read and learned individuals as Central Bankers. CBC just did a story about Goldman Sachs shuttling aluminum back and forth between warehouses to inflate the price and earn rent. That is micro's problem: RW communism.