A Teacher at the Front of Every Child’s Class: The Teacher Workforce and Demographic Shifts in Sub-Saharan Africa

• In Event: Teachers and Learning: Addressing learning poverty

Proposal

INTRODUCTION
There is no education without teachers. There will always be debates about teacher policy: how many teachers should we have per student, how much to pay teachers, how to support teachers, and how to hold teachers accountable for their work. But none of these debates eliminate the fundamental need for teachers in classes of students.

Demographic realities of both students and the current teacher workforce in countries across Sub-Saharan Africa mean that continuing to expand access to education—as proposed in the Sustainable Development Goals—and seeking to improve the quality of that education will likely require significant teacher recruitment in the coming years and decades. In Africa, student population growth is expected to be high through the next several decades, with most countries not peaking until after the year 2100. At the same time, different countries have dramatically different teacher age profiles, meaning that retirements and the need to recruit new teachers will peak at different points. This paper informs the demographic needs for teachers across the continent, and it proposes policy actions to meet those needs.

This paper projects teacher needs in various countries in Sub-Saharan Africa over the next 10-20 years. It combines projections of the student population at different ages, together with expected enrollment rates, the repetition rate, pupil-teacher ratios, and expected teacher attrition. The paper discusses the range of teacher needs across the continent. It also synthesizes existing literature on teacher hiring practices in the region (akin to the teacher hiring literature in Latin America, a la Cruz-Aguayo et al. 2017) and reviews analysis on teacher deployment (e.g., IIEP Pôle de Dakar 2016, Bashir et al. 2018 [Chapter 4]), with discussion of how those may affect teacher needs and the relevant policy options.

VALUE ADDED RELATIVE TO EXISTING LITERATURE
The two most significant cross-country efforts to predict teacher demand are UNESCO (2016) and UNESCO (2021). Both rely on UNESCO UIS data and UN Population Projections data. Both produce estimates of teachers needed, by schooling level, and by type (i.e., how many new posts versus how many to replace teachers who have retired or otherwise left the profession) and by country (in the case of the UNESCO 2021) or by region (in the case of UNESCO 2016). Teacher attrition is calculated simply: for example, in the UNESCO (2021) work, teacher attrition is calculated “from the most recent year with data available between 2015 and 2019,” assuming an attrition rate of 5 percent in cases where data were not available. Further, UNESCO (2021) model “assumes 100% enrollment in primary and secondary education.”

These assumptions may lead to significant overcounting or undercounting of teacher needs. In terms of teacher attrition, the age distribution of teachers varies significantly across countries. Recent analysis shows that average teacher age in Liberia (44.6 years) is nearly ten years older than average teacher age in Ghana (35.0 years) (Evans, Yuan, and Filmer 2022). Furthermore, different countries dramatically expanded their teacher workforce at different times, often coinciding with universal primary or secondary education policies. As a result, teacher attrition may reasonably be expected to be inconsistent both over time and across countries. Likewise, full enrollment in primary and secondary education is unrealistic in the coming years for most countries in Africa, even if free secondary education were implemented.

Our estimates relax both of these assumptions (teacher attrition and student enrollment). First, our project uses household survey data to understand the distribution of expected teacher retirements in coming years and will incorporate this into projections of teacher needs across countries. Second, rather than assuming full primary and secondary enrollment, we estimate two scenarios. In the first scenario, we project that enrollment rates continue to expand at rates similar to the rates at which they have recently expanded in these same countries. In the second scenario, we project that enrollment rates expand more rapidly—e.g., the scenario in which countries implement free secondary education policies—drawing on previous estimates of how rapidly enrollments have expanded in the wake of free schooling policies. We likewise include two scenarios on pupil-teacher ratios (PTR): in one, we assume the current ratio and in another, we assume another ratio more optimized for learning, informed by the current literature.

DATA
Data on the student population are available from the UN Population Projections. Basic educational data—e.g., repetition rates—are available from the UNESCO UIS data. Likewise, historical data on the expansion of enrollment rates is available from UIS data. The age distribution of teachers, which is crucial to understanding the upcoming pattern of retirement, is not available in the standard education data sources. However, we identified 20 household-based labor force surveys (mostly Living Standard Measurement Surveys) from 2014 and later. (More surveys may be available with further searching.) These types of surveys have been used to identify teacher age patterns in previously published work (Evans, Yuan, and Filmer 2022). Coverage of these surveys includes countries from southern Africa (Botswana, Namibia, South Africa), eastern Africa (Ethiopia, Tanzania, Uganda), western Africa (Liberia, Nigeria, Sierra Leone, Togo), and the Sahel (Burkina Faso, Mali). UIS data are available for other education variables for most of these countries; and we interpolate values in some cases if variables are missing but only for certain years.

RESULTS
This work is still in progress, but we will have a completed paper well before CIES. In initial results, we find that even with static enrollment rates (e.g., 80 percent primary enrollment in Liberia rather than assuming universal enrollment, which previous estimates do), there will be a massive increase in demand for teachers in the coming decades. For example, in Liberia we observe a 25 percent increase in need for teachers in ten years, driven roughly evenly by rising a population of school age and by increasing teacher retirements due to a high concentration of teachers ages 50-55 (i.e., likely retiring in 10-15 years).

Authors

• David K. Evans, Center for Global Development

• Fei Yuan, Harvard Graduate School of Education