Cover Image
close this bookTurbulence or Orderly Change? Teacher Supply and Demand in South Africa - Current Status, Future Needs and the Impact of HIV/AIDS (CIE, 2000, 36 p.)
View the document(introduction...)
View the documentMulti-Site Teacher Education Research Project (MUSTER)
View the documentList of Acronyms and Abbreviations
View the document1. Overview
View the document2. Background and Introduction
View the document3. The Demographic Characteristics of Teachers
View the document4. Incomes of Teachers and Non-teachers
View the document5. Teacher Turnover - The Dynamics of Teaching Employment
View the document6. Forecasting Basic Numbers
View the document7. Regionality and Micro-Regionality of the HIV/AIDS Epidemic
View the document8. Concluding Remarks
View the documentAnnex

6. Forecasting Basic Numbers

In this section we attempt a little crystal-ball gazing to forecast the basic supply and demand for teachers. Forecasting something so susceptible to social trends and policy shifts as teacher supply and demand is extremely hazardous. These sorts of forecasts have no rigorous confidence intervals; we cannot state our confidence in the results in the form that we are 95% certain that the forecast for the year t lies in the interval [x,y]. The demand side is relatively easy to forecast, but the supply side (and therefore the gap between supply and demand) is really quite chancy. All we can say is that these are not true forecasts, but simply conditional projections; that is, projections conditional on a whole host of assumptions, some of which have to do with likely demographic changes (including those related to HIV/AIDS), and some of which have to do with possible policy choices in the future.

The forecasts we offer are likely to be more reliable if we regard them as “broad brush” and medium- to longer-term in character and not short-term. For technical reasons beyond the scope of this report, projections are often particularly unreliable in the first few years. In the longer term, powerful determinants, such as basic demographics, whose parameters are more solid, tend to exert more influence than short-term phenomena such as teacher training or promotion policy.

Any forecasting depends on lists of assumptions which can be used to define different future scenarios. The process of systematically identifying assumptions itself helps gain insight into the likely behaviour of the main parameters which will affect teacher supply and demand. It also helps identify levels of confidence in assumptions and bands of uncertainty which can frame a range of projections to guide policy, rather than fruitlessly searching for a single “best” projection.

Annex Table A3 summarises the range of assumptions made in forecasting. Table 3 summarises the results in terms of likely gaps between supply and demand for different scenarios. If the gap is positive, this means that demand is greater than supply: there is a shortage. If the gap is negative, it means supply is greater than demand: there is a surplus.

Table 3: Conditional forecasts of gap between demand and supply


Explanation of scenario

Yearly gap 2001-2005

Yearly gap 2006-2010

Yearly gap 2011-2015

Demand (needed production) in mid-decade


Recent policy framework assuming declining enrolment due to early decline of population due to fertility transition, tightening of age controls, etc., and cutback in teacher training programmes, but no radical reaction by students choosing not to enrol in teacher training programmes. Assumed no HIV/AIDS impact known.






Same as scenario 1, but students react by radically choosing not to enrol, in response to lack of jobs due to relative hiring freezes in mid- and late-1990s; assumes that such a reaction is permanent. Assumed no HIV/AIDS impact known.






Assumes slower fertility transition (so no early population decline due to this factor), but worst-case HIV/AIDS scenario. No special attention to HIV/AIDS orphans via standard teacher training and supply. Assumes student over-reaction as in scenario 2.






Same as 3, but students’ choice of teacher training as a post-secondary option goes back up to nearly historic levels. Requires clear hiring messages from authorities, as well as low-cost study options for those choosing to be trained as teachers.






Similar to 3, but, on top, it is assumed HIV/AIDS orphans require special attention (L:E ratio of 10 to 1 for orphans). In addition it is assumed that student over-reaction is reduced by ½.






Similar to 5, but assumes that students’ choice of teaching as a post-secondary training option goes back up to the level of the early 1990s, or five times (400%) the current level





Source: calculated by the author.

Scenario 1 shows what would have happened had the policy assumptions on which the planning was based in the early 1990s turned out to be true. It indicates that the planning projections were fairly accurate given what was known at the time. It shows a small over-supply developing in the medium term after 2000. Considering that the teacher employment base is some 350,000, an apparent over supply of a few thousand is an acceptable result. An important assumption is that cutbacks in enrolment come faster than any increases in the graduation rate from programmes, or in the proportion of students who actually join the teaching profession. The small over supply appears later as demographic changes work through. Scenario 1 needs modifying in the light of what is now known about changes in key parameters.

Scenario 2 illustrates what seems to have happened as students reacted to poor prospects due to hiring freezes and general teacher labour market conditions in the mid 1990s. If students react to the poor hiring prospects and the closing of teacher training colleges by assuming that job prospects will be as bleak as they were in the late 1990s indefinitely, then recruitment will fall. Historically (in the early to mid 1990s), about 15% of matriculants chose to study teaching. By 2001 at best between 3 to 5% are choosing to enrol in teacher education. Even if an improvement in the graduation rate is assumed, and there is an improvement in the proportion of graduates who become teachers, there is a shortage starting now, which continues. However, the shortage is not as severe as one might think, because population growth is down, and enrolment in grade 1 is less bloated than previously.

Scenario 3 shows what may happen as HIV/AIDS affects the country’s teacher supply dramatically and students’ low propensity to enrol in teacher education continues. Very large yearly deficits in supply show up very early. Cumulative yearly gaps as high as 20,000 teachers show up by 2010.

In scenario 4 we assume that students recognise that hiring will not be permanently frozen. It is assumed that authorities send clear messages that hiring will resume (partly as a result of attrition related to HIV/AIDS). Possible changes in policy could also shorten the length of training or ease entry requirements. Eventually the shortage is addressed. But note that it might take some time to change student behavioural choices. If the messages are clear and strong, then the shortage could be addressed more quickly, instead of having to wait until 2006 or so. The scenario assumes only a very slow reaction by students.

Scenario 5 assumes that HIV/AIDS orphans require special attention, and that student propensity to enrol in teacher education doubles. That is, it is assumed that student choice of teaching as a post-secondary option is improved by 100%. A learner:educator ratio of 10:1 for HIV/AIDS orphans was used to generate this scenario. The costs of responding to special needs of HIV/AIDS orphans are immense. In this case a yearly shortage of some 38,000 and subsequently 48,000 teachers develops.

Scenario 6 is the same as scenario 5, but assumes student perceptions of teaching as a career improve dramatically and result in applications five times current levels. Despite this there is a yearly shortage of 12,000. This scenario is fairly important. It suggests that even in the extremely unlikely scenario that students go back to choosing teaching in the same proportions they did in the early 1990s, there would still be a very large shortage which is most acute in the near term.

In conclusion to this section, we believe the following cautious statements are justifiable.

· If the state attempts to pay special attention to the orphanhood generated by HIV/AIDS, and attempts to do so with formally trained teachers, then we simply do not see any scenarios that make this feasible, unless educational resources were to be heavily skewed towards this one task, and many other educational tasks dropped or minimised.

· If the state decides not to pay special attention to orphans, or admits that it simply cannot do so with formally trained and paid teachers, then it is possible to think of scenarios wherein the system could train sufficient numbers of teachers to cope with HIV/AIDS amongst teachers, and then address the issue of orphanhood through more informal and community-based means. Essentially, “all” that would be required is a return to the basic social parameters of transition between Grade 12 and teacher training that were common in the early 1990s. One would have to plan so that some 15% or so of matriculants would choose to become teachers, instead of the 2%-3% that are making this choice now. Our research suggests that this requires not so much a shift in pay policy or salary scales, as for the state to start sending out clear messages to secondary school students about the likelihood of jobs being available in teaching and improved planning and expanded enrolment capacity in cost-effective teacher education.

· The latter would require considerable analysis of options for the cost-effective training of teachers which is beyond the scope of this paper. It may suggest that teachers should not be trained through the relatively expensive, and perhaps relatively inefficient, methods common in the early 1990s and certainly common in the late 1990s. We calculate, for example, that the social cash cost (i.e., public and private, but without counting the private opportunity cost) of one year of pre-service training, by the late 1990s, was approximately R45,000 to R50,000 per student per year in teacher training colleges, and approximately R20,000 per student per year at higher education institutions. However, we note by the late 1990s, teacher training colleges were running well below capacity. When they were functioning at levels closer to capacity the cost would have been less.

· The issue of the real effectiveness of the training has to be faced. While some South African research suggests that pre-service teacher training is an effective form of expenditure by the state, it is unclear whether it is the subject-matter content of the training, or other aspects of the training are most efficacious. There is a risk that much training in the past may have been a form of screening for the profession with a limited amount of value added, or worse from a public investment point of view, might simply have been a stepping stone to other careers.

· Three to four years of pre-service training is likely to cost R60,000 R100,000 per teacher. Scenario 4 above, which is what is needed to reach some form of balance, suggests training some 30,000 new teachers per year. At a cost of R80,000 per teacher (to take the mid-range), the price tag is about R2.5 to R3 billion per year. Such an expenditure must be cost-effective. Before embarking on the sort of scale-up suggested above and in Scenario 4 as a way of coping with the effects of HIV/AIDS, we would suggest that this matter be addressed as a matter of utmost urgency.