I have already written three posts for "think at the margin, but why?". I have raised doubts about the usefulness of marginal thinking. I think the marginal thinking is useful but it may not be as useful as advertised by some textbooks. Also, the examples used to illustrate the usefulness of the marginal thinking are often not consistent with the mathematical part of this thinking.
The shortcoming of the examples used in textbooks is that they often involve sequential decision but usually decision is not made sequentially. We won't wait and see if one more can of coke should be purchased, having drunk one already. We won't stop the machine and assess if producing one more product is worthwhile. Sometimes the examples are too artificial and so cannot convince me very much. Yet, marginal thinking has its value and can't be ignored. I am simply not very satisfied with the examples given.
Recently I've encountered a news report about enrolling more non-local students for local universities. The news report spent much time on discussing the cost issue. The information it gave is roughly this. Local students pay $40,000 to $50,000 a year for their university education. Non locals pay $120,000 to $140,000. But universities are subsidized by the government for the operation. The government subsidy to universities amounts to $240,000 to $250,000 per student. The news report therefore said that enrolling more non-local students is loss-making as the student fee is lower than the average subsidy (or the cost of enrolling a student).
Of course, enrolling local students is also loss-making. But local students' parents perhaps pay taxes that are used to finance their children. Non-local students' parents do not pay taxes as local parents do and so there is no reason to enroll them at a fee below cost. That's the major contention of this news report.
In fact, I heard such a similar argument many years ago. In fact, every time there is a plan to enroll more non-local students, a similar argument will appear.
Why do I mention this in this blog? Answer: I think exactly this is a mistake due to common people's failure to think at the margin. Why "think at the margin" is important? Not because there are artificial examples to show its importance. It is because common people will really ignore it, and not once, but many times. The non-local student cost problem exactly illustrate this.
What is the mistake? It is about the cost of enrolling a student. Yes, on average, the cost may be as high as $240,000. But this is the average cost. Why the average cost is high? Because there is a high fixed cost for providing university education, including maintaining a team of high-quality university teachers. But these costs are largely fixed. Once the team is employed, the cost will not increase by enrolling, say, 1, 100, 1000 or 10,000 more students.
Thus, when assessing if enrolling more non-local students is loss-making or not, please don't look at average cost. Please look at marginal cost. If universities enroll 10,000 more non-local students, of course there is a marginal cost for doing so. But it won't be $240,000 for one more student. It will be much lower as $240,000 is only the average cost, which contains a bulk of fixed cost. If we think at the margin, we will find that the marginal cost of enrolling one more student is almost zero as universities will not provide any new facility and employ new staff to serve just one new student. The cost is not $240,000! Of course, the marginal cost of enrolling 10,000 more students is not zero. But it won't be $240,000 x 10,000 either. It must be much lower!
So, please, please think at the margin. This principle is indeed important.
