Principles of MOOConomics, part II: Elasticity of demand

My first post on MOOConomics generated a lot of discussion in the Microeconomics Principles class forum (it got more page views in five days than the entire CELT blog would in a month). I followed it up with a post about the highly elastic demand for MOOC content–or, what I thought was highly elastic demand.

Over the past few days, I’ve been thinking about the supply and demand curves for MOOCs. Both are pretty flat (i.e., elastic), so that it might be more accurate to talk about an “equilibrium price locus” more than an “equilibrium price point.” I think there’s good reason to think that whether the number of students enrolled is 10,000 or 100,000, the equilibrium price is virtually zero. Coursera is obviously willing (for now) to supply content indefinitely without regard for price. The price elasticity of supply approaches infinity.

What about demand? According to data for a course in bioelectricity offered through Coursera by Duke University, the number of students who watched the video lectures declined through each week of the course (see the first chart on page 8). The decline was steep at first, but then leveled off, probably as a core of very dedicated students kept watching them. This gives a hint as to what the demand curve would look like, if we plotted the quantity of course content demanded against the student’s time investment. This time investment can represent an opportunity cost of getting access to the course content.

I used data from the second chart on page 8 (also here) and made some assumptions about the amount of time it would take a typical student to reach each of the thresholds listed on the chart and came up with a possible curve for the aggregate demand for course content, as a function of opportunity cost (in hours). The course’s instructor estimated that it would take 6-8 hours per week to complete the course.

enter image description here

Just from the looks of it, demand is highly elastic until the opportunity cost reaches about 10 hours or so, and then becomes quite inelastic (elasticity ≈ -0.2 between the two leftmost data points). In other words, most students are very responsive to opportunity cost–they will not give up even fifteen minutes in order to gain course content. If Coursera decided to charge for access (that is, if its supply curve began to have a positive slope), then the equilibrium price would rise above zero, and it is safe to assume that MOOCs would suddenly not be so massive (or open!).

There are a couple things I am not entirely sure about, and I’d appreciate anyone’s help in sorting this out for me:

  1. Even though the tail of this demand curve looks highly inelastic, when you work out the numbers, it is in fact very inelastic (-0.2 to -0.5), because you are dividing the percentage change in quantity demanded by a very large percentage change in price. If the price goes from 0 to 1, it’s an infinite percentage change, so you get perfect inealsticity. My question is how we should understand price elasticity when we’re dealing with very inexpensive goods. Does what we learned last week still hold true?
  2. In working up the chart, I assumed that opportunity cost is a good proxy for monetary cost. Maybe that’s true, but maybe it isn’t. I can imagine that if the credential Coursera offered for completing a course became widely accepted, then the demand curve would become more linear. I can also imagine that a lot of the people who were willing to put in a lot of time would not be willing to put in a lot of money. We’re dealing with a strange market here.