On Saturday, I teased a forthcoming post about whether "Eat your pizza or aggregate demand collapses" is a good argument to have with less-than-completely-hungry children at Chuck E. Cheese's. The rest of the conversation will appear later this week, after my Principles of Macro and Intermediate Macro students have debated the merits and demerits of expansionary policy. Nonetheless, our trip raised an interesting question in light of last Tuesday's visit from West Virginia University's Andrew Young. Andy made a point that I remember Pete Boettke making in other contexts: Keynesian theory leads to Keynesian data; as he argues, the development of national income accounting proceeded on the basis of Keynesian assumptions about what should go where.* This complicates the job of one who might wish to run an empirical horse race between Keynesian, Austrian, and other business cycle theories to see which factors--malinvestment, collapsing aggregate demand, real shocks, regime uncertainty, financial fragility a la Hyman Minsky, I'm sure you can think of a dozen or so others, and here are some LearnLiberty videos in which Tyler Cowen explains competing theories--are most important for explaining business cycles. Andy's visit and what he said about business cycles and measurement made me move "re-read NBER conference volumes 24 and 30 very carefully" closer to the top of my to-do list.
Measurement is obviously problematic. The BEA counts "education" as "consumption spending" (see Cowen & Tabarrok's macro principles book, p. 103), for example, and this is definitely true for some students and some degrees. If people are in school to acquire human capital, then schooling should probably be counted as investment.
Here's a question I put to my intermediate macro students yesterday: when I take my kids to Chuck E. Cheese's, am I consuming or am I investing? It seems like "consuming" is the obvious answer. However, just as schooling can be used to build human capital, the time and energy we spend with our kids builds social capital. Presumably, this will pay off in higher future non-market consumption (a better relationship with my kids as such) and perhaps higher future market consumption (my kids are also potential future business partners). Measurements in which investment is classified as consumption and vice versa run the risk of being misleading.
Simplification is always and everywhere inevitable, differences in classification may not matter that much in a $14 trillion economy, and the national income accounts are good enough for government work. Things are definitely not always what they seem, though, and just as economics in the Austrian tradition takes an expansive view of human action, it should also take an expansive view of categories like "consumption" and "investment." Indeed, there are probably several great dissertations waiting to be written on the theory and history of economic measurement. There's also probably a grad student who is reading this right now and wondering what he or she should write about. If that's you, then here's a topic for your next paper. Get to work and change the world.
*-A qualification: this is not necessarily a bad thing. Indeed, I hesitate to write "Keynesian" because I've heard it used in a pejorative sense far too often.