The Economics Network

Improving economics teaching and learning for over 20 years

Teaching Relevant Economics (Keynote, DEBE 2005)

DEBE 2005 Keynote Speech

Prof. Monojit Chatterji, University of Dundee

Throw out the IS-LM model and Slutsky decomposition and start your syllabus from a blank page. Make the course relevant to student experience, employer needs and current research, says Monojit Chatterji in his provocative keynote speech. We are grateful to Prof. Chatterji for help adapting it for print.

I've been interested in this area for some time because I work in a system in which we compete for students with other subjects. The problem is students don't commit firmly to a subject when they come as freshers, they can walk in and out at any time. Effectively, we in Economics compete for students in an "internal market" in which bread and butter depends on holding onto your students. At the same time you have an obligation not to dumb down and I feel strongly about this too. The clear problem is how you retain students in this framework. Now a lot of what I have to say is geared for the ninety percent of students who may not be convinced that economics is for them. They are doing it as part of another programme; it's a requirement, European Studies or something of that sort and many of our students are of this kind. It's a similar problem to the "Internal Market" problem which those of us who teach in Scotland face.

Now for the ten percent students who are committed to economics, probably go on to do MScs, and PhDs the traditional way we have taught economics seems to have worked fine and I see no reason to change that. But really my purpose here today is to focus on the rest; on the vast majority of "floaters" for whom this method does not seem to have worked.

The first things we have to think about are choice of topics, the order of topics and the delivery. All three are inter-related as I will show but that is typically the order down which we have to think. I'm going to leave out issues of assessment and so on for now.

I'm going to start with each of these in turn.

Choice of topics

In my opinion, the single biggest source driving choice of topics is tradition. Think of how most of us teach a course particularly those who are young or new to a department and the Head of Department says, "Bloggs I want you to teach senior year macro which is a core paper, core subject" and you think "Oh my god I've never taught macro before but what can I say? I am the new person so I have to say I can do it". So what do you do? You do what everybody does and look at last year's syllabus and fine tune it and that completes your syllabus. In effect, no radical change is possible in this system. Serially correlated errors are perpetuated as yesterday's mistake becomes today's mistake, and the mistake lives forever because no one is in a position to change that.

So the questions to ask before you start on syllabus design are:

  1. Who are your students? Are they in the 90 or 10% I mentioned previously?
  2. How do I change this syllabus, given that students need to be motivated?

And the important decision in the choice of topics is not what you put in but what you don't put in. It is really tempting to try and be comprehensive in coverage. But why is this seen as a binding necessity? And where can one get advice on topic selection? A good starting point for this is employers. The traditional places we expect economists to be working in are in the Financial Sector, the Treasury and Government Economic Service generally. These employers will probably look to the 10% of committed students who are potential MSc seekers. But there are other employers who require the services of economists in a non-specialist environment. These employers may be a very useful source of information.

So the people I have asked about choice of topics are employers of this kind. I asked Evan Davis, Economics Editor of the BBC, (who gave the first keynote address at DEBE in Bristol). Many economists now working in the media. What skills do they need? As an example he mentioned present value calculations. Now this is a skill but one apparently that he found useful and that other people found useful in the context of his work. Whilst present value calculations are taught in most programmes, the amount of time devoted to it may well be insufficient for students to attain the fluency expected. The other person I have spoken to a lot on this is the former editor of The Economist. Now these are the kind of people who are really useful and we should encourage contact with others like them especially as they will be the ones interviewing our students for jobs. Does the Network have a mechanism for talking to a wide range of potential employers as to what they think students should know?

And that must inform what we teach. The choice of topics shouldn't be restricted by tradition which is a bad reason for doing anything in my view. Look at any syllabus or textbook from all over the world and there will be a stunning similarity. For example in a standard micro course we cover scarcity, resource allocation, competition, producer theory, monopoly, oligopoly, general equilibrium.

Ordering of Topics

Most textbooks follow a completely standard sequence of topics. Consumer theory, followed by Producer theory, perfect competition followed by general equilibrium and welfare and then right at the end a spirited dash through inter-temporal choice, monopolistic competition etc. There is no reason for doing it in this way, in this order at all, but we do. The so called extra topics - inter-temporal allocation, monopolistic competition, labour supply - are done at lightning speed and no one understands a word of it so many subjects just get lost or rushed. So the order in which you run your course matters a lot.

All of us are more ambitious than perhaps we should be and end up with more topics than we can finish. It can be true that the order in which you decide to run your course matters a lot. Re-arranging the order of a standard micro course and moving inter-temporal choice further up the tree instead of Welfare economics might mean that students have a deeper understanding of present value compared with Pareto.

The order of topics may actually help understanding as well. Monopoly and monopolistic competition and then competition as a limiting case is not the traditional order for teaching at all. It seems to me much easier to teach this way and it has more resonance with students. I think because they find this idea of the market setting the prices very difficult to cope with because every experience of theirs tells them the opposite. You walk into any store and there are prices and many people who work in stores have gone round putting the stickers on and you tell them in the ideal form of economic competition which is perfect competition nobody sets the prices, the market sets the price. It just does not ring true and it doesn't. We live in a world of price setting. What is competition? Competition is a situation where the price setting ability is so restrictive that price setters all end up choosing the same prices.

I can't remember the time in thirty years of teaching where I've come to the end of the course and said "Oh my god, I've got a spare lecture at the end so I can give a revision lecture on the whole course". It's never happened to me once so unless you guys are much better than I am then I'm sure you find yourself in the same position.


Once again the traditional way of delivering curricula is to start with some complicated general models and then we have a little example. And that again is a perfectly proper way to proceed, and has worked very well with motivated students but does not work so well with students who are struggling as it takes so long and so by the time you get round to the interesting parts, mainly applications, you have lost them. They are falling asleep or gone off to do something else - probably Business Studies.

Against that we should consider the use of approaches common in many other subjects - the case study method. It is increasingly used by us too, but still has not become a genuine alternative to the traditional method of theory first, example later. A colleague of mine once said he intended to "use economics of regulation to teach industrial economics" but when he was going to do this at undergraduate level he was going to teach them theory first and then talk about the case. But when he was teaching MBA students he was going to reverse that order; the case study was going to come first. Why this distinction?

What are the comparative advantages of these two approaches to delivery? The advantage of "case study first, economic analysis after" is that one has a much better chance of capturing the students' interest. A case study does not have to be complicated. By contrast, most of the examples that follow a piece of textbook theory appear to be quite artificial. It is much easier if you start the other way. Write down everything about the case study and ask "How much economic theory do I need to explain this?" and you will be amazed at how much you can strip out.

There is of course a downside to this. The case study method can leave big gaps in knowledge. No case study can possibly encompass all the concepts we need. So they must be used with care. As a generalisation, I would imagine that case-study-based approach will need to blend in with the traditional theory-first approach, with the balance tilting towards the latter as we move up the years from freshers to final year undergraduates.

I never would have thought that there's a connection between Agatha Christie and economics but my kids told me: it's just about speculation and you can unravel the whole thing once you put the missing facts together. This an example of how I would use a case study: tell the story, then extract what you can in terms of economic theory from it. You can't hope that any one case will cover everything: that's completely unrealistic so you have to find a balance. There's nothing to say you shouldn't teach a course which is half case studies and half driven in some other way. By this time you have hopefully got somebody interested and you can be a little bit more brutal in terms of banging in theory without them asking why the theory matters.

I teach to first-years things that people used not to see until graduate school. Ten years ago, how many people came across a full treatment of the efficiency wage in an undergraduate programme? They didn't typically, but I was able to do that, and with a very high success rate. A lot of my students did very well in it. It looked like an idiosyncratic first year of course, and it was. There were all kinds of first-year topics they didn't do. I made no apologies for that: I defined my learning outcomes to be exactly what came out of my two case studies.

So this requires a great deal of departmental support and co-operation. Whoever came on the back of me had to know that there were all these holes that they would have to fill up along the way. But what matters is that the hole gets filled over the course of the student's entire programme. Otherwise, you do the same stuff four times, with ever-increasing orders of mathematical complexity. As someone said about exams, from first-year to master's level the questions are the same; the only difference is the answer. I don't think this is a very useful way to teach.

Be radical: don't be frightened to go with something. Be prepared to kill a few sacred cows. I know what I am going to kill in my micro course: I am not going to teach Slutsky income and substitution effects: there is not going to be one word about them in the whole of thirteen weeks. I'd like someone in this room to tell me when they last used Slutsky in any research they did. I haven't used it in any of the papers I've ever published. So why do we teach it? Is it because John Hicks writes about it in "Value and Capital" and because the mathematics is so beautifully elegant. It seems to me that all of these are lousy reasons to teach it to undergraduates. It's going, for sure and there may be other sacred cows I'm going to kill. I don't want a student of mine going to a job interview, being asked "What did you learn in Economics?" and answering "I learned all about Slutsky" as if that's guaranteed to get you a job.

I haven't been asked to teach Macroeconomics yet, but sooner or later in a small department it's going to come my way. I'm going to use Geoff Wyatt's book, which is nice and easy and all about causation, which I like. I'm only going to drop one thing from that: why do we teach the IS-LM curve? Which bank controls the money supply, anywhere in the world? For fifteen years at least the Bank of England has been setting a minimum lending rate. Why don't we teach that and cut out a layer of unnecessary complication? Because of John Hicks. He is my favourite economist: If I were to sleep with an economics book under my bed, it would be "Value and Capital", but I still won't teach Slutsky or IS-LM. In some sense, IS-LM is almost worse than Slutsky. I think the IS-LM analysis is actually misleading whereas Slutsky is not misleading, just unhelpful.

IS-LM analysis is actually misleading because you tell students that the interest rate is determined endogenously but they read in the newspaper that a group of good wise men and true are setting an interest rate. Any good student will ask, "How can they set both money supply and the interest rate? How is that possible?" And furthermore, if you write down a formal macro model in which the interest rate is the endogenous variable, not the money supply, that model is actually different from the IS-LM model. In Geoff Wyatt's model, for example, it would actually behave differently. So you are actually telling them a more truthful story if you don't teach them the IS-LM, and it's actually easier because there is one loop in the causation process you are taking out. So you win twice: it's easier and it's truthful. And it increases student understanding of discussions in the newspapers.

So, be radical; throw out every syllabus you've ever seen and start with a blank page. It's really quite revealing and liberating. Think about the usable skills that students need, and you go a long way. Our syllabi will start changing. We'll all start looking a little different as we all start coming up with different case studies. So have fun, and don't do what you did yesterday!