≡ Menu

A Timely Note On Supply and Demand Analysis

David Henderson, over at EconLog, does a very nice job at exposing some of the flaws in Noah Smith’s criticism of Econ 101.  Here’s a comment (with typo corrected) that I left on David’s post:

David: Nicely done.

I’m sure that most people who teach principles of microeconomics do what I do in class when I explain the effects of price changes on quantity supplied and quantity demanded: I offer, in the course of this discussion, an analysis of elasticity. (Having once used Mankiw’s text in my classes, I know that he does so, too.) I make clear to my students three points that are relevant here:

First and most trivially, supply and demand analysis – as is true of all analysis – requires judicious use of ceteris paribus assumptions.

Second, supply and demand analysis itself tells us only about the direction of changes and not about the magnitude of changes. “A rise in the price of apples reduces the quantity of apples demanded each month” is a typical statement that is correctly made using supply and demand analysis. I make crystal clear to my students that in order to discuss the magnitude of the resulting fall in quantity demanded requires an understanding of own-price elasticity of demand. It is careless for Noah Smith to suggest that the possibility of highly inelastic demand as a feature of reality renders supply-and-demand analysis wrong, misleading, or unhelpful.

Third, when discussing elasticity I – like, I’m sure, most other econ-principles profs – explain that one of the determinants of elasticity is time. Specifically, both demand and supply are more elastic the longer the amount of time is available to adjust to price changes. I typically even use the minimum wage as an example, explaining that most of the jobs that are likely destroyed by a higher minimum wage are not destroyed immediately but, instead, are destroyed (or are not created) over time as employers have more time to adjust to the new, higher wage. And I do all this explaining not in a way that implies that differences in elasticity across outputs, across inputs, and across time somehow diminish the explanatory power of supply-and-demand analysis but, instead, as an integral part of making supply-and-demand analysis as explanatory as possible.

I add here that when I teach supply-and-demand analysis I always label the horizontal axis “Qty./t” – in order to make clear to my students that the quantities demanded and supplied are quantities demanded and supplied “during some specific period of time” such as “per day” or “per year.”  Changing “t” changes the elasticity of both demand and supply.  Any professor who teaches supply and demand in a way that implies that the elasticity of demand and the elasticity of supply are invariant to time, or who otherwise implies that time is not a relevant-enough factor to account for when doing supply-and-demand analysis, is a very poor professor indeed.  Such a professor is, as the always-insightful Jon Murphy suggests in another comment on David’s post, one who ought to refund to his students the tuition they paid for his faulty instruction.