6/2026: South Africa's population dynamics, economic growth, and employment prospects, 2030-2045

Read the full Occasional Paper by downloading the PDF below.

PUBLICATION INFORMATION

Title: South Africa’s Population Dynamics, Economic Growth, and Employment Prospects, 2030–2045

Authors: Shamsunisaa Miles-Timotheus and Pedzisai Ndagurwa, PhD

Publication type: Occasional Paper

Publication date: September 2026

PUBLISHED BY

Inclusive Society Institute | NPO Registration: 235-515

5 High Cape House, 21 Ben Avenue, Vredehoek, Cape Town, 8001, South Africa

PO Box 12609, Mill Street, Cape Town, 8010, South Africa

Tel: +27 (0)21 201 1589 | Email: info@inclusivesociety.org.za | Website: www.inclusivesociety.org.za

COPYRIGHT

© 2026 Inclusive Society Institute.

Unless otherwise expressly indicated, copyright in this publication is owned exclusively by the Inclusive Society Institute. All rights reserved. No part of this publication may be reproduced, stored, distributed or transmitted in any form or by any means without the Institute’s prior written permission, except as permitted by applicable law.

Brief extracts may be quoted for criticism, review, research or educational purposes, provided that the authors, title and publisher are appropriately acknowledged. Permission requests should be directed to admin@inclusivesociety.org.za.

SUGGESTED CITATION

Miles-Timotheus, S. & Ndagurwa, P. (2026). South Africa’s Population Dynamics, Economic Growth, and Employment Prospects, 2030–2045. Inclusive Society Institute.

USE OF ARTIFICIAL INTELLIGENCE

Artificial intelligence (AI) tools may have been used to assist with the drafting, editing, synthesis and presentation of this paper. All substantive content, interpretations and conclusions remain subject to human review and editorial oversight, and responsibility for the final publication rests with the authors and publisher.

DISCLAIMER

This publication has been prepared for research, discussion and public-information purposes. The views expressed are those of the authors and do not necessarily reflect the official position of the Inclusive Society Institute, its Board, staff, funders, partners or associated organisations.

While reasonable care has been taken to ensure the accuracy of the information at the time of publication, the Institute and the authors make no warranty regarding its completeness, accuracy or continued relevance. This publication does not constitute professional advice. To the fullest extent permitted by law, the Institute and the authors accept no liability for any loss or damage arising from reliance on its contents.

References to third-party organisations, products, services or websites do not constitute endorsement. The Institute is not responsible for the content, availability or accuracy of external websites.

SOUTH AFRICA’S POPULATION DYNAMICS, ECONOMIC GROWTH,

AND EMPLOYMENT PROSPECTS

2030–2045

by Shamsunisaa Miles-Timotheus and Pedzisai Ndagurwa (Ph.D.)

Abstract

This paper conducts a series of population projections for South Africa and use these projected populations as a base to estimate potential labour market outcomes, namely, unemployment rates, number of employed and the number of the unemployed. Population projections were based on the cohort component method (CCM), making use of a set of fertility, mortality and migration assumptions. The projections of labour market outcomes are conducted based on a set of gross domestic product (GDP) growth assumptions and Okun’s Law coefficients. The Okun coefficients, which measure the unemployment rate elasticity of GDP growth, included an empirically derived coefficient and three imposed coefficients. The results show that South Africa’s population and working-age population will continue to grow across all fertility and mortality scenarios, thereby substantially expanding the labour force.

Employment growth over the projection period will likely be insufficient to absorb new entrants to the labour market. Under the empirical Okun scenarios, unemployment remains persistently high, with only marginal differences observed between low (1%), moderate (2.5%), and high (4%) economic growth trajectories. Improved employment responsiveness leads to gradual declines in unemployment rates; however, even under the most favourable conditions, unemployment remains above 30% by 2045. Importantly, the number of unemployed individuals increases across all scenarios, reflecting the scale of labour-supply pressures.

The findings highlight that a weak relationship between economic growth and employment in South Africa will result in persistent high unemployment rates and point to the need for addressing the underlying structural nature of unemployment. While demographic change influences the scale of the challenge, it does not resolve the imbalance between labour supply and demand. The study concludes that addressing unemployment will require not only sustained economic growth but also structural reforms to enhance labour absorption.

Keywords: South Africa, Population projections, Employment projections, Unemployment, Economic growth, Okun’s Law, Cohort-component method, Working-age population, Labour absorption, Demographic dividend

1. Introduction

South Africa’s development path is influenced by both demographic change and economic performance. As a middle-income country facing high unemployment, understanding how these forces interact has become increasingly important for long-term planning. On the one hand, the country has experienced a sustained decline in fertility over recent decades, reflecting increased urbanisation, expanded access to education, and greater use of modern contraception. On the other hand, population momentum continues to drive growth in the absolute size of the population, particularly within the working‑age population. At the same time, South Africa’s economy has recorded persistently low growth rates since the early 2010s, limiting its capacity to generate employment. The coexistence of a growing labour supply and weak demand for unskilled and semi-skilled labour has resulted in one of the highest unemployment rates globally.

Most analyses of South Africa’s future labour‑market prospects examine either demographic change or economic growth in isolation, rather than integrating the two, as seen in OECD, (2025), van Aardt et al. (2026) and Statistics South Africa (2023). Population projections are typically used to anticipate future demand for services, infrastructure, and social protection, while employment projections often rely on macroeconomic assumptions that treat population trends as fixed or implicit (Buettner, 2022). This separation obscures the fundamental interaction between population dynamics and labour‑market outcomes. Population growth and age structure shape the supply of labour, while economic growth determines the economy’s ability to absorb that labour into productive employment. From a theoretical perspective, this interaction is central to the concept of the demographic dividend, which emphasises that favourable age structures yield economic gains under the right conditions (Bloom et al., 2003).

This paper seeks to bridge this analytical gap by integrating cohort‑component population projections with employment projections based on a dynamic version of Okun’s Law and a constrained imposed coefficient. By jointly considering demographic and economic scenarios, the paper provides a more comprehensive assessment of South Africa’s employment prospects over the medium to long-term. Specifically, the paper addresses the following questions: How will South Africa’s population and working‑age population evolve between 2025 and 2045 under alternative fertility assumptions? What will be the employment and unemployment rates for the given GDP growth scenarios?

The paper makes three main contributions. First, it provides transparent population projections based on explicit fertility assumptions, highlighting the uncertainty in South Africa’s demographic future. Second, it applies a dynamic Okun’s Law framework to project employment outcomes under alternative growth paths. Third, it combines these elements to generate a scenario matrix that links demographic and economic trajectories, offering policy‑relevant insights for long‑term planning.

2. Demographic and economic context

South Africa is situated in the later stages of the demographic transition, characterised by low fertility and mortality alongside continued population growth driven by demographic momentum. Total fertility has fallen substantially from levels above five children per woman in the mid‑twentieth century to around replacement level in recent years. Despite this decline, population growth has not halted. This is because the large cohorts born during earlier periods of higher fertility are now entering or already in the working-age years, creating both opportunities for economic growth and challenges in employment absorption.

The working‑age population (15-64 years) is particularly important for economic development, as it represents the potential labour supply. In theory, an increasing share of the working‑age population can generate a demographic dividend in a context with sufficient job creation, skills development, and supportive institutions. In practice, South Africa has struggled to turn this potential into meaningful economic gains. Structural constraints, skills mismatches, and slow economic growth have limited employment creation, resulting in widespread unemployment and underemployment.

From an economic perspective, South Africa’s growth performance since the global financial crisis of 2009 has been subdued. Average GDP growth has remained well below the levels required to meaningfully reduce unemployment, averaging 1.3% annually between 2009 and 2025 (World Bank, 2026). While short‑term cyclical factors have played a role, structural issues, including energy constraints, weak investment, and institutional challenges, have constrained growth. Consequently, employment growth has lagged labour‑force growth, particularly among young people.

The literature on employment dynamics often draws on Okun’s Law, which posits a negative relationship between economic growth and unemployment. While the strength of this relationship varies across countries and over time, it provides a useful framework for linking macroeconomic performance to labour‑market outcomes. In the South African context, studies have found that the relationship between GDP growth and unemployment is inconsistent (Chiranga et al., 2025). This is consistent with evidence across African countries, where the relationship between growth and unemployment is often weak or absent due to structural or labour market factors (Ibourk & Elayaoui, 2024). Nonetheless, Okun‑type relationships remain valuable for scenario analysis and long‑term projections.

3. Methodology I: Population projections

The gold standard method for projecting populations into the future is the cohort-component method (Ernst et al., 2023; Smith et al., 2013). The cohort-component method (CCM) is a well-established formal demographic technique for estimating future population sizes, taking into consideration empirically derived assumptions about fertility, mortality and migration (Preston et al., 2001). The procedure for CCM involves three main steps. First, starting with a current snapshot of the population, broken down by age and sex, we project to the next five years, by forward-surviving each age group by sex, to obtain the number of each cohort that will be alive by the fifth year. Second, we estimate the number of children who will be born alive to women aged 15-49 years over the next five years and add them to the population. Finally, we estimate the net migration for the population by subtracting the total number of people who move out of South Africa and adding those who will move into the country. The base population for the projections is the 2025 mid‑year population estimates for South Africa, disaggregated by five‑year age groups and sex. The projections extend to 2045, with results reported for the years 2030, 2035, 2040, and 2045.

3.1 Fertility assumptions

Three alternative fertility scenarios were specified to reflect uncertainty about future fertility trends, namely low-fertility, medium, and high-fertility. The low‑fertility scenario assumes a continued downward trend in fertility, with the total fertility rate (TFR) declining to approximately 1.0 by 2045. The medium‑fertility scenario assumes a more moderate decline, with TFR reaching approximately 1.8 in 2045, representing a realistic continuation of recent trends. The high‑fertility scenario assumes that the TFR remains constant at around 2.26, a situation known in demographic literature as a stalled fertility trend. Fertility declines are projected based on the current TFR of 2.26. To project future fertility, the current age-specific fertility rates (ASFRs) are scaled down until they are consistent with a target TFR while maintaining the age distribution of fertility. The sex ratio at birth is assumed to be 105 males for every 100 female births.

3.2 Mortality

The standard life table approach was used to derive the mortality experiences of the population of South Africa. The population data for the life table were obtained from the 2022 census, and the counts of deaths were obtained from the latest death registration data, which includes all 2022 deaths, including late registrations that were captured in 2023. It is noteworthy that the 2022 census had the highest undercount rates ever recorded in South Africa post-1994, and quality concerns were acknowledged by Statistics South Africa on some of its data, such as fertility. However, the census remains the only data source for national population data. Furthermore, the post-enumeration survey (PES) conducted shortly after each national census has been found to be effective in generating adjustment factors for calibrating final census counts (Gumbo, 2016). Three mortality scenarios were assumed for the projections. The first scenario was the current mortality level based on real data. This was considered a high-mortality scenario, based on 2022 deaths, a period affected by the COVID-19 pandemic. The second scenario assumed a medium mortality level, with improvements in mortality to a magnitude of 10% reduction in the number of deaths at each age from the current level. The third scenario assumed a low mortality level, assuming 20% reduction in the number of deaths at each age from the current level.

3.3 Migration

Migration is an important component of the population dynamics of South Africa, as it is one of the major destinations on the African continent for international migrants. Over the last intercensal period, 2011-2022, South Africa had an annual average net migration rate (NMR) of 2.9 per 1,000, that is, a net gain (arrivals minus departures) of roughly three persons for every 1,000 people in the country. Net migration was assumed to be constant over the projection period at approximately 56,000 persons per year, reflecting the estimated net migration level observed in the 2022 census. Most migrants entering South Africa are concentrated in the 20-39 age group, and this age distribution is assumed to remain constant throughout the projection period. In the projection process, net migration is incorporated after the population has been aged forward using survival method and after births have been accounted for.

4. Methodology II: Employment projections using

Okun’s Law

Employment and unemployment projections were conducted using Okun’s Law coefficient, which relates changes in unemployment to changes in gross domestic product (GDP), allowing for lagged effects. It is traced to Okun (1962) who observed an inverse association between unemployment and GDP over short-term periods. In his seminal work, Okun (1962) found that a three-percentage increase in GDP was inversely associated with a one-percentage point decrease in unemployment rate. Some studies conducted after Okun’s seminal work have also confirmed the inverse relationship between GDP shifts and unemployment (Prachowny, 1993; Boďa, & Považanová, 2025).  In Other studies, such as by Blackley (1991), this relationship was found to be closer to two-percentage points increase in GDP being associated with one-percentage point decrease in unemployment. Applications of the dynamic version of Okun’s Law specify an empirical GDP-employment relationship that allows for time lags and past trends, allowing for the quantification of the relationship in a temporal continuum (Aguiar‑Conraria, Martins & Soares, 2020).

In this study, the dynamic Okun’s coefficient was estimated using an Autoregressive Distributed Lag (ARDL) model. The dynamic Okun is an improvement on the static version which assumes only immediate employment responses to GDP shifts. To obtain the long‑run dynamic Okun coefficient, the sum of short‑run effects is divided by a persistence factor, calculated as one minus the coefficient on lagged unemployment, which accounts for the dependence of unemployment on its past values. The resulting measure represents the long‑run responsiveness of unemployment to GDP shifts, which captures a more structurally accurate understanding of the relationship between economic growth (or decrease, stagnation) and labour market outcomes over time. In this study, the dynamic Okun coefficient was estimated using GDP and employment data from the 2009-2025 period.

It is noteworthy that the empirically derived Okun coefficient for South Africa for the period 2009 to 2025, including the COVID-19 pandemic period, contradicted theoretical Okun assumptions, producing a positive coefficient of 0.0193. To mitigate the impact of the COVID-19 pandemic on the GDP-unemployment relationship, the empirical Okun coefficient was estimated using data for the 2009-2019 period, yielding a coefficient of -0.0841. This coefficient, however, represents a weak relationship between economic growth and unemployment reduction in South Africa, a situation observed in most low- to medium-income economies (COVID-19). Some of the reasons cited for the weak Okun coefficient in economies like South Africa’s are that unemployment is driven by structural factors and that GDP growth has been capital-intensive rather than labour-intensive (Habiyaremye et al., 2022; Moloto et al., 2025). Furthermore, GDP growth can stimulate a return to the job market by previously discouraged workers, adding to the additions from cohort shifts, which subdues changes in unemployment rates during times of economic growth (Dinkelman, & Pirouz, 2002; Ngundu & Ngalawa; 2023; Tchereni, 2021). To mitigate the weaknesses of an empirical dynamic Okun coefficient for South Africa and to test prospects of unemployment rates under different GDP-employment regimes, we also used three Okun coefficients (-0.02, -0.1, and -0.2) to capture diverse conditions.  Values around -0.02 capture scenarios of minimal employment response, -0.1 represents a typical weak response observed in emerging markets, and -0.2 reflects a stronger, but still plausible, labour market adjustment consistent with empirical evidence (Sovbetov, 2025).

Three GDP growth scenarios were specified: a low‑growth scenario of 1% per annum, reflecting prolonged economic stagnation; a moderate‑growth scenario of 2.5%, consistent with partial structural reform and recovery; and a high‑growth scenario of 4%, reflecting sustained reform and improved investment conditions. Employment trajectories were projected under each growth assumption, combined with population projections and a constant labour force participation rate to derive unemployment outcomes. The baseline for the labour force participation rate was derived from the 2025 Quarterly Labour Force Survey (QLFS) as an annual average, which was 59.5%.

The overall estimation procedure proceeded as follows: population projections were conducted by first deriving age‑sex population cohorts for each target year (2030, 2035, 2040, and 2045) using the cohort component method, incorporating survival ratios, age‑specific fertility rates, and age‑specific net migration, using 2025 as the base year. From these projections, the working‑age population (15–64) is aggregated for each period and converted to the labour force by applying a fixed labour force participation rate of 59.5%, the average for the base year 2025. Changes in the unemployment rate are then modelled using a constrained Okun relationship, where GDP growth scenarios of low (1%), medium (2.5%) and high (4%) were translated into marginal adjustments in the unemployment rate using one empirically derived (0.0193) and three imposed (-0.02, -0.1, and -0.2) Okun coefficients. The resulting unemployment rates are applied to the projected labour force to derive the number of unemployed, with employment obtained as a residual.

5. Results I: Population projections

Under all mortality and fertility scenarios, South Africa’s total population is projected to continue growing between 2025 and 2045, though at different magnitudes. Projections under low-fertility scenarios produced the slowest population growth and a gradual stabilisation of the working‑age population toward the end of the projection period. The medium‑fertility scenario yields moderate growth, while the high‑fertility scenario results in a substantially larger population and faster expansion of the working‑age group.

Age‑structure changes differ markedly across scenarios. Lower fertility accelerates population ageing and reduces youth dependency, while higher fertility sustains a younger age structure but increases pressure on education and labour markets. These dynamics have direct implications for employment prospects when combined with economic growth trajectories.

5.1 Based on high mortality, medium fertility and constant

migration trends

Based on high (current) mortality, medium fertility, and constant migration trends, South Africa’s population is projected to grow steadily from 63.1 million in 2025 to 73.6 million in 2045. The growth trajectory reflects ongoing demographic momentum. This scenario represents the most plausible pathway, assuming no major shifts in fertility, mortality and migration patterns.

Table 1. Projected population under high mortality and medium fertility

The age-sex structure of South Africa is projected to be dominated by adults in the 25-39 and 50-59-year age groups. Figure 1 below compares the 2025 age-sex structure with the projected one for 2045. There will be just as many 0-9-year-olds as 50-59-year-olds in 2045 if mortality remains constant and fertility decline continues at the current trend.

Figure 1. Current mortality, medium fertility and constant migration

5.2 Based on high mortality, high fertility and constant migration

trends

Assuming fertility remains constant at current levels, population growth is more pronounced, reaching approximately 77 million by 2045. Compared to the baseline year, the higher population size reflects the cumulative effect of sustained fertility, which reinforces the expansion of younger cohorts and contributes to a larger overall population over time.

Table 2. Projected population size under a high mortality and high fertility scenario

With high mortality and fertility remaining high (constant at TFR = 2.26), the age distribution of the population in South Africa will shift upward, with an increasing proportion of the population aged 45 and older. The proportion below 20 years will decrease, but not to the same extent as would occur when fertility decreases (as is the case in Figure 1). Figure 2 below compares the age-sex structures for 2025 and 2045 under high mortality and high fertility.

Figure 2. Age-sex structure between 2025 and 2045 in the event of high mortality and high fertility

5.3 Based on high mortality, low fertility and constant migration

trends

In a low-fertility scenario, in which the total fertility rate declines to 1, population growth slows considerably, reaching 67.9 million by 2045. While the population continues to grow in the short- to medium-term due to demographic momentum, the long-term trajectory indicates gradual stabilisation, with significantly smaller cohort sizes emerging over time.

Table 3. Population projections under a high mortality and low fertility scenario

Under a high mortality and low fertility scenario, South Africa’s age structure will undergo a significant transition by 2045. The age pyramid will have a narrow base that broadens in the youth and adult age groups, as shown in Figure 3 below.

Figure 3. Prospective age-sex structure shifts under high mortality, low fertility

5.4 Based on medium mortality, medium fertility and constant

migration trends

Under medium mortality assumptions and current fertility trends, the population is projected to increase to 74.6 million by 2045. Improvements in mortality contribute to population growth, although the overall trajectory remains similar to the baseline scenario.

Table 4. Population projections under the medium mortality and medium fertility scenario

Assuming continuation of current fertility trends (medium fertility scenario) and a 10% decline in mortality (medium mortality scenario), South Africa will undergo a moderate transition towards an ageing population structure. As shown in Figure 4 below, the proportion of cohorts under 15 years will be smaller than in 2025. As in other scenarios, the 45-64 age group will increase their share of the population.

Figure 4. Medium mortality and medium fertility

5.5 Based on medium mortality, high fertility and constant

migration trends

With both improved survival rates and sustained fertility levels, population growth is further amplified, reaching approximately 78 million by 2045.

Table 5. Population projections under the medium mortality and high fertility scenario

A high fertility regime combined with medium mortality will cause notable shifts in the age structure [Figure 5 below], but to a lesser extent compared to a medium fertility regime.

Figure 5. Age-sex structure under medium mortality and high fertility

5.6 Based on medium mortality, low fertility and constant

migration trends

When fertility declines to below replacement level. Under medium mortality conditions, population growth moderates, reaching 68.8 million by 2045. Although improved mortality supports population increases, the impact of reduced fertility ultimately slows overall growth.

Table 6. Population projections under medium mortality and low fertility

A combination of medium mortality and low fertility will yield a population in 2045 where children below 15 years of age will account for a small proportion of the population [Figure 6].

Figure 6. Age-sex structure under medium mortality and low fertility.

5.7 Based on low mortality, medium fertility and constant

migration trends

In a low mortality scenario, the population is projected to reach 75.6 million by 2045. The increase reflects the growing number of individuals surviving into older age groups, alongside sustained fertility trends.

Table 7. Population projections under low mortality and medium fertility

The base of South Africa’s age structure will narrow with medium fertility and low mortality, with defined gains in the 50-64 age range [Figure 7].

Figure 7. Age-sex structure under low mortality and medium fertility.

5.8 Based on low mortality, and high fertility

This scenario produces the highest population growth, with the total population reaching approximately 79 million by 2045. The combination of low mortality and high fertility leads to substantial expansion across all age groups, reinforcing long-term population growth.

Table 8. Population projections under the low mortality and high fertility scenario

As shown in Figure 8 below, high fertility will slow the transition to an ageing population pyramid compared to medium fertility.

Figure 8. Age-sex structure under low mortality and high fertility.

5.9 Based on low mortality, low fertility and constant migration

trends

Even under low mortality conditions, a decline in fertility to below-replacement level results in slower population growth, with the total population reaching 69.8 million by 2045. While increased longevity sustains growth in older age groups, reduced fertility limits expansion in younger cohorts.

Table 9. Population projections under the low mortality and low fertility scenario

As shown in Figure 9 below, low mortality and low fertility will result in an age structure with a very narrow base, just as in medium and high mortality scenarios involving low fertility.

Figure 9. Age-sex structure under low mortality and low fertility.

5.10 Summary of population projections

The population of South Africa is set to continue growing, as seen across all scenarios. The low mortality-high fertility scenario will produce the fastest population growth, while the high mortality-low fertility scenario will produce the slowest growth. The most likely to prevail is the medium mortality, medium fertility scenario. We anticipate that fertility rates will continue the downward 2008-2025 trend (medium fertility scenario), and mortality rates will fall to pre-COVID levels (medium mortality scenario). The likely population estimate for South Africa in 2045 is therefore 75 million.

Figure 10. Summary of population prospects by projection scenarios

The proportion of the population under 15 years will shrink, although the absolute number will increase. As the current (2025) cohorts in the age groups 0-19 and 25-39 years, to 15-44 and 40-59 years, respectively, they will account for the majority of the population, signifying a transition to a mature population age structure. This has important implications for services and policies such as those aligned to post-secondary education, health for an ageing population.

6. Results II: Employment and unemployment

projections

Unemployment projections were conducted based on observed demographic events from population data, and these were the scenarios of high mortality, medium fertility, and constant net migration. The sections below, therefore, present the unemployment projections by type of Okun coefficient and GDP growth rate.

6.1 Empirical Okun coefficient and 1% GDP

Over the projection period, labour market outcomes deteriorate gradually despite increases in employment. The unemployment rate declines steadily from 32.4% in 2025 to 31.6% by 2045.

Although employment increases steadily from 16.1 million in 2025 to 19.9 million in 2045, this growth is insufficient to keep pace with the expanding labour force. As a result, the number of unemployed individuals increases significantly, from 7.7 million to 9.2 million over the same period.

Overall, the simultaneous rise in both employment and unemployment highlights a key structural challenge: while jobs will be created with modest GDP growth, they will not be generated at a rate fast enough to offset labour force growth. This leads to persistently high and increasing unemployment, reflecting weak labour-market absorption.

Table 10. Unemployment projections based on a dynamic empirical Okun coefficient and 1% GDP growth

6.2 Empirical Okun coefficient and 2.5% GDP growth

Under a moderate economic growth scenario of 2.5%, labour market outcomes improve marginally compared to the low-growth baseline. While employment increases slightly faster, reaching 20 million by 2045, this remains insufficient to absorb the expanding labour force, which grows to 29.1 million. As a result, unemployment remains persistently high, with slight improvement from 32.4% in 2025 to 31.4% in 2045, only marginally lower than in the 1% growth scenario. The number of unemployed individuals increases to approximately 9.1 million. Overall, the results indicate that moderate economic growth will have a limited impact on the levels of unemployment, reinforcing the existing weak relationship between employment and economic growth in South Africa.

Table 11. Unemployment projections based on a dynamic empirical Okun coefficient and 2.5% GDP growth

6.3 Empirical Okun coefficient and 4% GDP growth

Under this scenario, labour market outcomes deteriorate slightly over time, reflecting the economy’s limited capacity to absorb a growing labour force. The unemployment rate decreases gradually from 32.4% in 2025 to 31.3% by 2045. Employment increases steadily from 16.1 million to just over 20 million over the projection period, indicating continued job creation. However, the number of unemployed individuals still rises in absolute terms, from 7.7 million to slightly over 9.1 million. Overall, the results suggest that even 4% annual GDP growth in the context of the currently existing relationship between economic growth and employment will not be sufficient to reduce unemployment rates meaningfully.

Table 12. Unemployment projections based on a dynamic empirical Okun coefficient and 4% GDP growth

6.4 Low Okun coefficient and 1% GDP growth

In this scenario, labour market outcomes show slight improvement over time, although overall conditions remain challenging. The unemployment rate declines marginally from 32.4% in 2025 to 32.1% by 2045, suggesting a modest strengthening in the economy’s ability to absorb labour.

Employment increases steadily from 16.1 million to 19.7 million over the projection period. However, the number of unemployed individuals also rises, from 7.7 million to 9.4 million, due to continued labour force growth.

Overall, the results suggest that while employment growth can be realised with a low Okun coefficient, there will be no meaningful impacts on the unemployment rate, and the size of the unemployed population will also grow.

Table 13. Unemployment projections based on a low Okun coefficient and 1% GDP growth

6.5 Low Okun coefficient and 2.5% GDP growth

This scenario reflects a gradual improvement in labour market outcomes over time. The unemployment rate declines slightly from 32.4% in 2025 to 32.0% by 2045, suggesting a modest increase in the economy’s capacity to absorb labour.

Employment grows steadily from 16.1million to 19.8 million, indicating continued job creation over the projection period. However, despite this positive trend, the number of unemployed individuals still increases in absolute terms from 7.7 million to just over 9.3 million, driven by ongoing labour force expansion.

Overall, while the declining unemployment rate points to incremental improvements in labour market performance, the persistent rise in the number of unemployed highlights the scale of the challenge. Job creation, although improving, remains insufficient to fully counterbalance the growth in labour supply, underscoring the structural nature of unemployment.

Table 14. Unemployment projections based on a low Okun coefficient and 2.5% GDP growth

6.6 Low Okun coefficient and 4% GDP growth

This scenario shows a slightly greater improvement in labour market outcomes over time. The unemployment rate declines from 32.4% in 2025 to 31.9% by 2045, indicating a gradual strengthening in the economy’s ability to absorb labour.

Employment increases consistently from 16.1 million to 19.8 million, reflecting steady job creation throughout the projection period. Despite this improvement, the number of unemployed individuals still rises in absolute terms, from 7.7 million to 9.3 million, due to continued labour force expansion.

Overall, while the declining unemployment rate suggests incremental progress, the simultaneous increase in the number of unemployed highlights the scale of labour supply pressures. Even under improved conditions, employment growth remains insufficient to fully absorb the growing labour force, reinforcing the persistence of structural unemployment.

Table 15. Unemployment projections based on a low Okun coefficient and 1% GDP growth

6.7 Medium Okun coefficient and 1% GDP growth

This scenario reflects a more noticeable improvement in labour market outcomes over time. The unemployment rate declines steadily from 32.4% in 2025 to 31.4% by 2045, indicating a gradual strengthening in the economy’s capacity to absorb labour.

Employment increases consistently from 16.1 million to just below 20 million, indicating stronger job creation than in previous scenarios. However, despite this improvement, the number of unemployed individuals still rises in absolute terms from 7.7 million to more than 9.1 million due to continued labour force growth.

Overall, while the declining unemployment rate suggests meaningful progress, the persistence of a large and growing unemployed population highlights the scale of the challenge. Even with improved employment performance, labour demand does not fully keep pace with labour supply, underscoring the structural nature of unemployment.

Table 16. Unemployment projections based on a medium Okun coefficient and 1% GDP growth

6.8 Medium Okun coefficient and 2.5 % GDP growth

This scenario shows a continued improvement in labour market outcomes over time. The unemployment rate declines from 32.4% in 2025 to 31.3% by 2045, indicating a gradual strengthening in the economy’s ability to absorb labour.

Employment grows steadily from 16.1 million to almost 20 million, reflecting sustained job creation across the projection period. Despite this progress, the number of unemployed individuals still increases in absolute terms from 7.7 million to approximately 9.1 million, due to the ongoing expansion of the labour force.

Overall, while the declining unemployment rate suggests improving labour market performance, the persistence of a large unemployed population highlights the continued imbalance between labour supply and demand. Even under improved conditions, employment growth remains insufficient to fully absorb labour market pressures, reinforcing the structural nature of unemployment.

Table 17. Unemployment projections based on a medium Okun coefficient and 2.5% GDP growth

6.9 Medium Okun coefficient and 4% GDP growth

This scenario reflects the greatest improvement in labour market outcomes among those considered. The unemployment rate declines steadily from 32.4% in 2025 to 31.1% by 2045, indicating a gradual but more pronounced strengthening in the economy’s ability to absorb labour.

Employment increases consistently from 16.1 million to just over 20 million, representing the highest level of job creation across the scenarios. Despite this improvement, the number of unemployed individuals still rises in absolute terms from 7.7 million to approximately 9 million, driven by continued labour force expansion.

Overall, while the declining unemployment rate points to improved labour market performance, the persistence of a large unemployed population underscores the scale of the challenge. Even under the most favourable economic growth conditions, unemployment will not be reduced markedly.

Table 18. Unemployment projections based on a medium Okun coefficient and 4% GDP growth

6.10 High Okun coefficient and 1% GDP growth

This scenario shows a more substantial improvement in labour market outcomes over time. The unemployment rate declines steadily from 32.4% in 2025 to 30.5% by 2045, indicating a stronger capacity of the economy to absorb labour compared to previous scenarios.

Employment increases significantly from 16.1 million to 20.2 million, reflecting robust job creation throughout the projection period. Despite these gains, the number of unemployed individuals still rises in absolute terms from 7.7 million to approximately 8.9 million, driven by continued labour force growth.

Overall, while the declining unemployment rate suggests meaningful progress and improved labour absorption, where the GDP-employment relationship has greater elasticity. However, the persistence of a large unemployed population highlights the pressure from labour supply expansion, which suggests that higher GDP growth is imperative.

Table 19. Unemployment projections based on a high Okun coefficient and 1% GDP growth

6.11 High Okun coefficient and 2.5% GDP growth

This scenario reflects continued improvement in labour market outcomes over time. The unemployment rate declines steadily from 32.4% in 2025 to 30.4% by 2045, indicating a strengthening in the economy’s ability to absorb labour.

Employment increases significantly from 16.1 million to 20.3 million, demonstrating sustained job creation across the projection period. However, despite these gains, the number of unemployed individuals still rises in absolute terms from 7.7 million to just above 8.8 million, due to ongoing labour force expansion.

Overall, the results suggest greater employment utility of increase GDP growth to 2.5% per annum compared to 1%. Nonetheless, 2.5% GDP growth will still be insufficient to reduce the number of unemployed in the country.

Table 20. Unemployment projections based on a high Okun coefficient and 2.5% GDP growth

6.12 High Okun coefficient and 4% GDP growth

This scenario represents the greatest improvement in labour market outcomes across the projections. The unemployment rate declines steadily from 32.4% in 2025 to 30.2% by 2045, indicating a notable strengthening in the economy’s capacity to absorb labour.

Employment increases substantially from 16.1 million to 20 million, reflecting sustained and relatively robust job creation over the projection period. Despite this progress, the number of unemployed individuals still rises in absolute terms, from 7.7 million to just below 8.8 million, driven by growth in the labour force.

Overall, while the declining unemployment rate signals improved labour market performance, the persistence of a large unemployed population highlights the ongoing pressure from labour supply expansion. Even under the most favourable economic growth conditions and a more elastic employment-GDP relationship, unemployment is set to remain above 30%.

Table 21. Unemployment projections based on a high Okun coefficient and 4% GDP growth

6.13 Summary of unemployment projections

Projections of unemployment rates in South Africa for given Okun coefficients and GDP growth rates reveal important patterns [Figure 11 below]. Firstly, any increase in the GDP growth rate will reduce the unemployment rate, but in most cases only marginally. If the current unemployment elasticity of GDP growth (measured by the empirical Okun coefficient), unemployment rates will fall marginally to a lowest possible 31.3% in 2045 from 32.4% in 2025. If the unemployment elasticity of GDP growth weakens, unemployment rates in South Africa will be worse than they are now. Improvements in the unemployment elasticity of GDP growth will lead to more defined reductions in unemployment rates (high Okun coefficients). The results imply that South Africa needs to address structural factors driving unemployment in the country. Addressing the structural factors of unemployment will improve unemployment elasticity of GDP growth, such that even modest economic growth can result in greater decreases in unemployment.

Figure 11. Summary of projected unemployment rates

7. Discussion and policy implications

The findings of this study highlight a persistent and structurally embedded disconnect between population dynamics, economic growth, and labour market outcomes in South Africa. Across all scenarios, the working-age population and labour force expand substantially over the projection period, driven by demographic momentum. As shown in the population projections, even under declining fertility assumptions, population growth continues in the medium term, reinforcing sustained pressure on the labour market.

The employment projections reveal that expanding labour supply will lead to increasing numbers of jobless individuals if the economy does not create sufficient jobs due to the momentum of population growth. Under the employment projections based on the empirical dynamic Okun coefficient, even a relatively strong 4% economic growth is insufficient to meaningfully reduce unemployment rates. Instead, unemployment remains persistently high, with only marginal differences across 1%, 2.5% and 4% GDP growth scenarios. This finding suggests that the relationship between economic growth and employment in South Africa is weak, and in some cases, near-neutral. This projected situation has characterised the South African context over the past three decades, with studies documenting employment contraction in primary sectors like agriculture, forestry and fishing, and mining and quarrying, and demographic momentum as underlying drivers (Bhorat & Oosthuizen, 2008; Hodge, 2009).

The application of constrained, partially imposed Okun coefficients provides further insight into the employment prospects for South Africa. Scenarios with higher employment responsiveness show gradual declines in unemployment rates. However, even under the most optimistic assumptions, unemployment remains above 30% in 2045. Importantly, while unemployment rates decline, the absolute number of unemployed individuals continues to increase across nearly all scenarios. This reflects the scale of labour supply growth and highlights the limitations of relying solely on economic growth to solve unemployment.

The relationship between demographic and economic factors is central to the projection outcomes. High fertility scenarios amplify labour market pressures by increasing the size of future cohorts entering the labour force, thereby requiring significantly higher levels of job creation to stabilise unemployment. Conversely, low fertility scenarios reduce labour supply pressures but do not eliminate the unemployment challenge, particularly under conditions of weak economic growth. These findings reinforce the argument that the labour market challenges for South Africa cannot be resolved through favourable demographic trends.

Overall, the results point to a future characterised by population growth and persistently high unemployment rates in South Africa. The persistence of high unemployment, even under improved economic conditions, reflects underlying issues such as capital-intensive growth and skills mismatches, which limit the economy’s labour absorption capacity. The findings of future population and unemployment projections have several important implications for economic and labour market policies in South Africa, which are discussed below.

First, the results indicate that economic growth alone will be insufficient to address unemployment in South Africa, with the continuation of the current GDP-unemployment relationship suggesting a worse outlook. To attain a favourable employment outlook in the face of an unavoidable demographic momentum, South Africa needs to promote labour-intensive primary and secondary sectors such as agriculture, forestry, and manufacturing, as well as render meaningful support to small and medium-sized enterprises, which have greater potential for job creation. By deepening the economic policy’s focus beyond a growth-centric approach, the country will develop labour market resilience that is less prone to global economic shocks, thus maintaining strong employment conditions.

Second, there is a need for targeted labour-market interventions to better align labour supply and demand. The persistence of high unemployment alongside increasing positive GDP growth suggests that skills mismatches remain a critical constraint. The policy shift from polytechnic-type post-school education in South Africa to TVET that became primarily concerned with teaching and instruction at the expense of applied technical functions accentuated the skills mismatch in the labour market (Kraak, 2018).  It is particularly noteworthy for South Africa that the political and economic restructuring that characterised the transition from apartheid to a democratic dispensation, where government institutions were being privatised and technikons being converted to universities, severely damaged the Design, Engineering, Entrepreneurship and Management (DEEM) function that traditionally implied closer alignment between training and industry skills needs (Kraak, 2008). The strong cooperative relationships previously built up between the state-owned enterprises, the technical colleges and the technikons to train and skill artisans, technicians and technologists were scaled down and, in many instances, permanently lost (Kraak, 2008). It is therefore imperative that South Africa looks to it’s past to make relevant investments in education to better align the workforce with the economy's needs.

Finally, the findings underscore the importance of adopting a coordinated policy approach that integrates demographic and economic planning. The interaction between population dynamics and labour market outcomes suggests that policies should not be developed in isolation. Instead, a holistic strategy that considers population trends, economic growth, and labour market conditions simultaneously is required to effectively address unemployment.

8. Conclusion

This study conducted population projections by sex and five-year age groups. The projected population sizes of the 15-64 age range were used as a basis to estimate future unemployment rates, using a combination four Okun coefficients and three GDP growth rates. The four Okun coefficients, which measure the unemployment elasticity of GDP growth, were empirical (based on real past data for South Africa from 2009 to 2019), low coefficient (which represent the weakest GDP-unemployment relationship), medium, and high (strongest possible GDP-unemployment relationship in South Africa) coefficients. The results from population projections suggest significant population growth by 2045. The most realistic projected population size by 2045 is 74.6 million, based on the assumption that mortality will decrease from the 2022 level (which was high because of the residual COVID-19 pandemic effects), and that fertility will continue to decrease at the same steady rate as observed between 2022 and 2024. The unemployment projections revealed potential positive labour market outcomes, although only marginal. Should the relationship between GDP growth and unemployment rates that has prevailed in South Africa continue up to 2045 (empirical Okun coefficient), the country will experience marginal decreases in the unemployment rate, to between 31 and 32%. The number of the employed will increase for any GDP growth rate, but so will also be the number of the unemployed. The growth in the numbers of the unemployed will be driven by the momentum of population growth affecting the working age group.  Even under declining fertility assumptions, labour supply expands significantly, placing sustained pressure on the labour market.

While employment increases across all scenarios, the pace of growth remains too slow to absorb new entrants into the labour force. As a result, unemployment remains persistently high, with only marginal improvements observed even under more favourable economic conditions. Notably, higher GDP growth does not translate into substantial reductions in unemployment, reinforcing the weak responsiveness of employment to economic growth in the South African context.

The analysis further shows that variations in demographic assumptions shape the scale of the challenge but do not fundamentally alter its nature. Higher fertility amplifies labour market pressures, while lower fertility moderates them; however, neither scenario resolves the structural imbalance between labour supply and demand. Similarly, stronger employment responsiveness improves outcomes but remains insufficient to significantly reduce unemployment levels.

Overall, the findings highlight the relevance of the argument that South Africa’s unemployment challenge is structural rather than cyclical. Economic growth alone is unlikely to generate the scale of employment required to absorb the expanding labour force. Instead, addressing unemployment will require a combination of sustained economic growth, greater employment intensity, and targeted structural reforms to enhance labour-market absorption. Education reforms may be one avenue to address the structural barriers to employment. Furthermore, growth in the labour-intensive sectors is also crucial if South Africa is to realise significant labour market gains.

This study contributes to the literature by demonstrating the importance of jointly analysing demographic and economic dynamics when assessing future labour market outcomes. Linking population projections with employment scenarios, it provides a more comprehensive understanding of the scale and persistence of unemployment in South Africa. Future research should focus on refining estimates of country-specific Okun coefficients and on exploring sector-specific employment dynamics to further strengthen the evidence base for policy interventions.

References

Aguiar‑Conraria, L., Martins, M. M. F. & Soares, M. J. 2020. Okun’s law across time and frequencies. Journal of Economic Dynamics and Control, 116, 103897.

Bhorat, H. & Oosthuizen, M. 2008. Employment shifts and the ‘jobless growth’ debate. In A. Kraak & K. Press (Eds.), Education, employment and skills in South Africa: Human Resources Development Review 2008:50-68). Pretoria: HSRC Press

Blackley, P.R. 1991. The measurement and determination of Okun’s law. Evidence from state economies. Journal of Macroeconomics, 13(4), 641-656.

Bloom, D., Canning, D. & Sevilla, J. 2003. The demographic dividend: A new perspective on the economic consequences of population change. Santa Monica: Rand.

Boďa, M. & Považanová, M. 2025. A quarter century of Okun’s law in scholarly literature. Journal of the Knowledge Economy:1-56.

Buettner, T. 2022. Population Projections and Population Policies. In: May, J.F., Goldstone, J.A. (eds) International Handbook of Population Policies. International Handbooks of Population,11. Springer, Cham. https://doi.org/10.1007/978-3-031-02040-7_21

Chiranga, N. Zhanje, S. & Chingoiro, S. 2026. An empirical insight into the validity of the Environmental Kuznets Curve hypothesis in the SADC region. Front. Environ. Econ. 4:1708624. https://doi.org/10.3389/frevc.2025.1708624

Dinkelman, T. & Pirouz, F. 2002. Unemployment and labour force participation in South Africa: A focus on the supply-side. Economic Research Southern Africa Working Paper.

Ernst, J., Dräger, S., Schmaus, S., Weymeirsch, J., Alsaloum, A. & Münnich, R. 2023. The influence of migration patterns on regional demographic development in Germany. Social Sciences, 12(5):255.

Gumbo, J. D. 2016. Undercounting controversies in South African censuses. Johannesburg: University of the Witwatersrand.

Habiyaremye, A., Habanabakize, T. & Nwosu, C. 2022. Bridging the labour market skills gap to tackle youth unemployment in South Africa. The Economic and Labour Relations Review, 33(4): 786–805.

Hodge, D. 2009. Growth, employment and unemployment in South Africa. South African Journal of Economics, 77(4): 488-504

Ibourk, A. & Elaynaoui, K. 2024. Policy lessons from Okun’s law for African countries. International Review of Applied Economics, 38(5):564-593.

Kraak, A. 2008. Incoherence in the South African labour market for immediate skills. Journal of Education and Work, 21(3):197-215 https://doi.org/10.1080/13639080802214050

Kraak, A. 2018. The shifts to tertiary technical and vocational education and training and the demise of South Africa’s former ‘Technikon’ system. Journal of Vocational, Adult and Continuing Education and Training, 1(1). https://doi.org/10.1080/02692171.2024.2337348

Moloto, A. N., Ramasimu, N. F., Motsei, L. L. & Muchie, M. 2025. Graduate unemployment, skills mismatch, and the dynamics of labour mobility in South Africa: A systematic literature review. International Journal of Applied Research in Business and Management, 6(5). https://doi.org/10.51137/wrp.ijarbm.369

Ngundu, M. & Ngalawa, H. 2023. A sectoral analysis of output elasticity of employment in South Africa. South African Journal of Economic and Management Sciences, 26(1), a4825. https://doi.org/10.4102/sajems.v26i1.4825.

Obst, T. 2022. Dynamic version of Okun’s law in the EU15 countries - The role of delays in the unemployment‐output nexus. Scottish Journal of Political Economy, 69(2):225-241.

OECD. 2025. OECD Economic Surveys: South Africa 2025. Paris: OECD Publishing. https://doi.org/10.1787/7e6a132a-en

Okun, A.M. 1962. Potential GNP. Its measurement and significance. In: Yale University (Eds), Proceedings of the business and economic statistics section of the American Statistical Association: 98–104. Cowles: Foundation. https://milescorak.files.wordpress.com/2016/01/okun-potential-gnp-its-measurement-and-significance-p0190.pdf

Prachowny, M. F. 1993. Okun's law: theoretical foundations and revised estimates. The review of Economics and Statistics: 331-336.

Preston, S. H., Heuveline, P. & Guillot, M. 2001. Demography: Measuring and modeling population processes. Blackwell Publishers

Smith, S. K., Tayman, J. & Swanson, D. A. 2013. A practitioner's guide to state and local population projections. Springer Science & Business Media

Sovbetov, I. 2025. The dynamics of Okun’s law: Cross-country analysis across economic cycles (1980–2023). Empirica. https://doi.org/10.1007/s10663-025-09662-2

Statistics South Africa. 2023. Labour market dynamics in South Africa, 2023. Pretoria Statistics South Africa.

Tchereni, B. H. M. 2021. Nonlinear modelling of economic growth and unemployment in South Africa. Economica, 16(6):273-282.

Van Aardt C., Meiring J., Tustin D., van Tonder J., Hugo j., Jhatham A., Moyo A. & Letebele L. 2026. UASA South African employment report and 4th UASA/BMR Employment Index. UASA-BMR-2026_FIN-Digital-File.pdf

World Bank, 2026. DGP growth (annual %) - South Africa. https://data.worldbank.org/indicator/NY.GDP.MKTP.KD.ZG?end=2024&locations=ZA&start=2010