Noah Smith has some good ideas on this, and the CORE project (here is a presentation at INET’s annual conference) should have a new curriculum by the end of this year. But the reactions of many will echo Noah’s: there is just no room for any new stuff. It is certainly true, speaking about the macro component, that there is a danger we teach much too much material at this level. Some of what we teach appears contradictory: like the AS curve and the Phillips curve.
So my first point, which I have made before, is that we can get rid of a lot of stuff that is simply out of date. Like the LM curve (and theories of money demand that go with it). And the Aggregate Demand curve which is derived from it. And Mundell Fleming which is an open economy version of it (and inconsistent with UIP to boot). And the money multiplier (which, apart from being very misleading, is unnecessary if we stop fixing the money supply). But why not really get this bonfire going? Do we need to teach the Keynesian multiplier? As there are good reasons to think that the closed economy government spending multiplier (with a given level of real interest rates) is around one, what is the point?
Of course a lot of this would come back, in some form, in a more advanced macro course. However I have always thought the acid test for what should be included in an introductory course is whether it is something that a person who studies no more economics really needs to know. I would submit that all of the above fail this test.
What has to stay in? The IS curve of course: monetary policy is all about using interest rates to control aggregate demand. However I agree with John Cochrane that this should be based on the two period consumption model (which students with large loans can relate to), and not investment theory. The Phillips curve is central to how pretty well everyone thinks about macro, so that has to be there. It can be taught as an empirical regularity, introducing the macro history of the 1970s at the same time.
Sometimes people have told me that you need to say something about money if you want to talk about Quantitative Easing (QE). I think exactly the opposite is true: QE shows up how ridiculous the LM curve stuff is. QE represents a huge increase in bank reserves - and the money supply hardly moves (thank you money multiplier). How much simpler, and more realistic, to just talk about short and long interest rates. Dispensing with money allows us to spend time talking about the zero lower bound, and events since the financial crisis. I would use this to motivate a discussion of fiscal policy and debt.
Would I replace the LM curve with a ‘monetary policy curve’, expressing preferences over inflation and output, or a Taylor rule? I’m tempted not to, because when you do something like this, students stop thinking about monetary policy as a choice. The example I sometimes use is an accidental (not countercyclical) temporary fiscal expansion that is foreseen. So many students let that increase output and inflation, and then have a Taylor rule react. But of course if the fiscal shock is known, any sensible monetary authority would attempt to completely counteract the impact of that shock.
What about the ‘supply side’. I agree with Mankiw’s text that we can treat labour supply as fixed at this level. Together with a medium term assumption of fixed capital gives us all we really need to motivate a Phillips curve with a natural rate. Deriving a labour demand curve from profit maximisation tells us that increases in labour saving technology do not lead to increases in unemployment, which is nice, but it has the cost of confusing students (and policymakers) when we subsequently assume demand determined output.
I would replace Mundell Fleming with a combination of a net export function (which gives us a relationship between aggregate demand and competitiveness) and Uncovered Interest Parity (UIP). A key idea that should be taught at this level is that in a small open economy, it is the real exchange rate and not the real interest rate that ensures aggregate demand equals supply in the medium term. I think introductory macro should also say something about fixed exchange rate regimes.
So there you have it. Econ 101 with just three basic relationships: an IS curve, a Phillips curve and UIP. I would use some of the space created to talk about basic issues and common confusions, like the relationship between price flexibility and output gaps, or between involuntary unemployment and wage flexibility, or why Says Law does not hold and why the General Theory got written. Comments welcome on anything else that really should be in there.