There has been a delightful discussion going on between Phil Miller, John Palmer and John Chilton over a professor's constituency, particularly as it relates to economics professors teaching the first course their majors will take. Many argue that students are consumers, but Palmer dismisses that idea:

I like the concept of student as customer, except that in a market explicitly based on asymmetric information, it makes sense for students to pay professors to make/encourage/induce the students do something they would not otherwise do. Students are an important constituency, but catering to their current wishes is probably not a very good idea since profs know the subject (and one might hope) more about how to teach it than the students do. In this sense, teaching evaluations by students should probably have no more than a 10% weight on overall assessments of professors.

Chilton argues that the professors in the upper-division courses are the customers who receive students trained by the principles instructor. Miller wonders if it isn't the student's employers after graduation. Palmer, though, echoes my thoughts in touting the benefits of principles in and of itself.

Paul Heyne made his name as a teacher, in fact, from not following this advice. And
although I'm no Paul Heyne, one thing I know is that there are many aspects of the Economic Way of Thinking that I must teach my students because if I don't do it, they won't learn them in their upper-level courses. In other words, if I take the upper level courses as a parameter, then I could teach nothing but technique and math and really give my students a leg up for those courses. I don't think that would be doing them a great service, though, and it completely ignores those students who will not be taking any more courses in the subject.

There's a debate within my own department over this very question right now. As you might take it from this post, I'm in agreement with Palmer. Heyne is a wonderful book that teaches much economics but will be of little use to prepare students for the math in most intermediate theory texts. And in a school like ours, with 1100 of 1200 principles students taking no more economics, do you really want them spending much time learning constrained optimization at the expense of learning the nature of economic thinking?

Russell Roberts gave a lecture I heard once many years ago in which he summed up economics as three postulates, which he represented as acronyms:

1. SI -- people act as if motivated by their self interest.
2. NFL -- no free lunch, or more often known as TANSTAAFL.
3. MC=MB --marginal benefit = marginal cost, the notion that rational decision occurs at the margin. The corrolary I always teach is sunk costs or "don't cry over spilt milk", which is something Heyne does so well.

If I just send students armed with those three concepts, understood well, will they do well in intermediate theory? Probably not. Will they understand the principles of economics? My view is yes, but your mileage may vary. For a young professor it would be a hard decision without departmental support.

UPDATE (3/17): Phil suggests we add PRTI as a second corrolary to MC=MB.