5. Case study: Are government budget deficits always bad?
‘Are government budget deficits always bad?’ is one of the threshold network exercises which was used on the first year Introductory Macroeconomics course at Coventry University, following teaching on the Keynesian model. The exercise was used with three separate seminar groups. At the beginning of each seminar, the students were allocated to groups of 4–6 students. The discussions which took place within each group of students as they completed the exercise were recorded. The discussions for two groups from each seminar, chosen at random, were transcribed and analysed. Although the lecturer was in attendance for the duration of the seminar, help was only offered when students got ‘stuck’ and only by asking students questions as opposed to giving answers.
The scenario is thus: ‘The country is currently in recession and this has led to lower tax revenue and higher expenditure. The result is a large budget deficit. The government decides to raise taxes and lower government expenditure. Is the government’s decision a good idea?’ (This exercise is given in full in Appendix 3). Students were asked to identify, from a list, which threshold concepts might be useful in answering this question. The list consisted of the following concepts: inflation, interaction between markets, the multiplier (cumulative causation), investment, withdrawals, injections, social costs, scarcity, consumption and oligopoly. The students were also required to ‘Draw an appropriate diagram to illustrate the above scenario and comment’.
What the exercise uncovered in terms of understanding
Analysis of the discussions demonstrated that the difficulties students encountered and their progress could be categorised under two headings: (i) misconceptions and (ii) problems with modelling.
- The most common misconception or confusion was between the terms ‘investment’ and ‘government spending’. A less surprising confusion was between the concepts of money and income. However, for the purposes of this particular exercise, this was not crucial.
What was more surprising was the desire of many students to include social costs as a relevant concept. They rationalised this desire by discussing social costs in terms of the increased unemployment resulting from the recession. They also discussed social costs in terms of reduced government spending on education. The concept of social cost was not relevant to answering this question – this was an attempt by students to incorrectly try and make the concept relevant to the question, instead of correctly discarding it. Another concept which was not readily rejected was that of scarcity. This concept was incorrectly attributed to the scarcity of jobs in a recession.
There were also confusions with regard to Government Spending (G) and Taxation (T). Some groups assumed G = T, whilst another group initially equated Government Spending with unemployment pay!
(ii) The second major problem area was with regard to modelling the scenario. Most of the groups decided immediately it was either a good idea or a bad idea, without even considering a model. When students did eventually try to use a model, it was by trying to remember where particular lines went on a diagram, rather than by demonstrating any deep understanding of the model. Many groups tried to introduce elements not yet covered in the module (e.g. labour markets). Some groups were confused with regard to the possible impact of reduced injections and increased withdrawals on income.
Diagrams were often poorly drawn. The multiplier was not even mentioned or recognised as an important concept until it was noticed on the list! A great deal of time was ‘wasted’ trying to justify the use of concepts which were not relevant to the question.
Some groups got bogged down with trying to work out the nature of the tax, i.e. lump-sum or income related. In this particular question, it did not matter, but in all previous questions that the students had attempted it had mattered. This problem did not need the level of detail encountered in previous problems and students found it difficult to work without the framing given to them in the previous exercises which involved numbers and/or algebra. They found it difficult to transfer their skills from a very structured economic question to an economic problem couched in more general terms.
What was achieved
By the end of the seminar, all the groups had reached the correct conclusions, i.e. that it would be a bad idea to increase taxes and reduce government spending in a recession – some with a greater depth of understanding than others. After the students had completed the exercise, there was some class discussion and feedback. All students were also given a feedback sheet which identified the appropriate concepts and included a relevant diagram.
In conclusion, the concepts which students had the most difficulty with were Investment/Government Spending, Social Costs, Scarcity, Injections/Withdrawals and the multiplier. However, although students encountered misconceptions and followed the wrong path at various points, all the groups had some idea of the necessary concepts and the consequences of particular actions by the end of the seminar. The exercise exposed misconceptions that would otherwise have remained hidden and allowed incorrect understanding to be explored and rejected.
With regard to the modelling, there was far too much reliance on memory rather than understanding. However, it was definitely the case that students learnt by getting it wrong initially and then working out why it was wrong. It was good to make mistakes. One of the strengths was using this exercise as a group exercise. Members of the group often explained to each other why a particular line of thinking was right or wrong. It was very heartening to hear students talking about Economics problems to each other.
This particular exercise helped in understanding where students were having difficulties and what appeared to be the most prevalent misconceptions. It will certainly be used again along with many of the other exercises that have been developed as part of the Embedding Threshold Concepts project.