When teaching econ, start with the parts that work

Mike Konczal caused a splash when he suggested the other day that we're teaching econ backwards:

Today, first-year undergraduate students typically start with microeconomics, or the study of individuals and individual markets. This begins with the study of abstract, decontextualized, markets, where supply and demand work perfectly, individuals exist in isolation, and they effortlessly trade with others in isolation of society, the law, and politics... 

In their second class, students begin to learn macroeconomics, or what happens when you add up all those markets... 

What if macroeconomics came first, before the study of individual markets? If were to reverse the typical curriculum, the first thing undergraduates would encounter wouldn’t be abstract theories about people optimizing, but instead the idea of involuntary unemployment and the idea that the economy could operate below its potential... 

[I]t turns out that things like institutions, regulations, income distribution, the way markets are introduced into formerly Soviet countries, and the viciousness of the business cycle and mass unemployment all actually matter.

Paul Krugman endorsed the idea.

I pretty strongly disagree with this suggestion. I think that if business cycle theory were the main purpose of economics, then it might have merit, but this is not the case.

Macroeconomics, and particularly business cycle theory (the theory of why we have recessions), is econ's glamour league. Macro is so important that we elevate it to an equal plane with micro in graduate programs - there's an "introductory micro sequence" and an equally sized "introductory macro sequence". Out there in the real world, everyone and their dog wants to know why recessions happen and how we can prevent or escape them.

But despite its glamour, macro is really only one small part of economics. In addition to the causes of recessions, economics investigates the following:

* how tax policy influences business conditions and human welfare (tax econmics)

* why workers get and lose jobs and are paid the wages they are paid (labor economics)

* why we see the patterns of international trade and capital flows that we see (international econ)

* why people make the consumption, saving, investing, and other economic decisions that they make on a day-to-day basis (decision theory, behavioral economics, experimental economics)

* how auctions work (auction theory)

* how people can optimally play various games and other competitive endeavors (game theory)

* how companies compete and get profit (industrial organization)

* why financial asset prices move the way they do, and how investors should and do behave (financial economics)

* how government should provide public goods (public finance)

* why cities and industrial clusters form and grow and decline (urban economics)

* why some poor countries get rich (development economics)

* why some rich countries experience bursts of productivity growth (growth economics)

* how to use statistics to measure all this stuff (econometrics)

...and more!

So you see, there is a LOT more to econ than business cycle theory. Sure, business cycles interact with most of these things, but the things all interact with each other. Business cycle theory has assumed a special place among the constellation of econ fields, but only because A) people get really upset about recessions, and B) business cycle theory is very contentious and controversial.

But here's the thing: Business cycle theory is contentious and controversial precisely because it doesn't work very well. Cursed with uninformative data, macroeconomists have so far failed to definitively answer the following Big Questions:

1. What causes recessions?

2. Can recessions be forecast in advance?

3. Can recessions be prevented by government policy, and if so, how?

4. Why do inflations happen?

5. Why do hyperinflations happen?

6. Why do "jobless recoveries" happen, and can they be prevented?

This doesn't mean macroeconmists have done nothing. They have catalogued a vast array of business cycle facts and features. They have identified a few policies whose effect can be predicted fairly reliably (e.g., if the Fed raises interest rates it will cause a sharp recession in the short term). They have debunked a lot of old, dangerous ideas about what the central bank should do.

But if we teach kids macro first, we will be deliberately starting them off on the stuff that doesn't really work. Notice that no other science does this. Take an intro physics class, and it starts with Newton's Laws, projectile motion, classical electrodynamics, and optics - not the deepest stuff, not the cutting edge, but the stuff that has worked really, really well for hundreds of years. The successes. Only five or six years later do you get to the string theory and the cosmology and the speculative, cutting-edge stuff.

Similarly, I think we should start kids off with the econ theories that work. What are the econ theories that work? Well, there are a lot, but here are a few:

* Supply and demand work for many, many things. If a storm destroys your orange crop, the price of oranges will go up. The invention of a popular new orange juice will cause the price of oranges to go up. And so on. Many of these are so obvious that we take them for granted - they don't use fancy math, and the fate of nations doesn't hang in the balance. But these are real successes for economics.

* Game theory works very well in some situations, such as auctions, where everyone knows the payoffs and the rules.

* Tax economics works very well when predicting the impact of certain kinds of taxes (less well for others).

* Financial economics gets a great many things right. When interest rates go up, bond prices go down. Diversification of stocks can reduce risk without reducing expected return. Etc.

* Public finance theory works when explaining many externalities. By modeling pollution as a negative externality, economists managed to create some very effective "cap and trade" systems.

* Econometrics has had a lot of successes in predicting stuff, and it's never too early to start learning the baby versions.

Actually, there are a lot more. This was just a quick short list off the top of my head.

These things are often simple. They are often un-glamorous. People don't argue about them a lot. But they have led to real improvements in humanity's ability to control our world. In other words, they are scientific successes. I think we should teach these things before we start teaching about problems that economists have been butting their heads against, like business cycles or development econ.

(And actually, we mostly do teach these things first, at least at the undergrad level. So would I keep Econ 101 the same? No, I'd emphasize empirical evidence more than we do. Just like intro physics classes make you do labs and verify the equations of projectile motion, I'd have undergrads take a look at data and do some regressions to at least get an idea of how well some of these theories match reality.)

Addendum: I also think that undergrad macro, when we do teach it, should be taught from the standpoint of "Why we don't get what's going on." Teach the Loanable Funds model, but then show how poorly it's performed. Etc.