Abstract
The need for a comprehensive understanding of the relationship between demographic dynamics and economic performance is pressing, given the significant demographic changes that have taken place in post-apartheid South Africa, as well as the socio-economic challenges confronting the country since the dawn of democracy in 1994. This entails changes in population growth rates, structural age changes, migration, and urban populations. The population of South Africa has been increasing due to factors such as natural increase and immigration, especially from other African countries. However, COVID-19 has impacted migration and service delivery, thus affecting demographic patterns and South Africa's growth. To achieve its objective of assessing the link between demographic and economic performance in post-apartheid South Africa, the study conducts four empirical analyses. These analyses highlight the importance of demographics for favorable economic outcomes. Thus, it provides enabling strategic policy tools that could help policymakers leverage demographic benefits for sustainable growth.
The thesis's first empirical chapter examines the effect of demographic dynamics on household savings in pre-COVID-19 South Africa using data from 1995 to 2019 across all nine provinces. The study employs the panel autoregressive distributed lag (PARDL) and Dumitrescu-Hurlin (D-H) panel causality methods. Unlike previous studies, and given South Africa's unique population groupings, this study focuses on the effect of South Africa's working-age population, which is disaggregated into the four main racial groups (Asian/Indian, Black, Coloured, and White) on household savings in South Africa for the period from 1995 to 2019. South Africa comprises nine provinces, each with four major race groups and varying household saving rates. This analysis is expected to yield valuable insights into how key demographic variables influence household savings across South Africa’s nine provinces. Thus, the findings regarding the impacts of the working-age population of the four racial groups on household savings across South Africa's nine provinces, using provincial-level data, will contribute to the body of knowledge on the topic. The empirical results of the PARDL procedure used in this investigation are presented in two distinct models. The empirical results of the two models reveal evidence of a long-run (cointegration) relationship between demographic dynamics and household savings in South Africa in both models.
The second empirical analysis of the study explores the impact of economic growth on urban population across the eight metropolitan municipalities in South Africa, using a balanced
v
annual panel dataset for the period 1995-2019. The study utilises the panel data technique, complemented by the threshold technique, in investigating the relationship between economic growth and urbanisation in South Africa. The empirical findings of the study show that there is a long-run equilibrium relationship (cointegration) between economic growth (GDPG) and urban population growth (URBG) across the eight metropolitan municipalities in South Africa. The threshold analysis indicates that there is a negative linear relationship between economic growth (GDPG) and urban population growth (URBG) at the significance level of 1 percent (%) signifying a threshold level of 4.3972. This relationship becomes significant after GDPG falls below the threshold value of 4.3972 (4.40%).
The third empirical study assesses the impacts of demographic factors on economic growth (GDPGR) in both the long-run and short-run, using the PARDL technique on South African provinces from 1996 to 2022. One point of difference from the earlier studies is that this study selectively examines four underlying facets of population dynamics: namely, net immigration growth (NMGR), birth growth rate (BGR), death growth rate (DGR), and population growth rate (POPGR). The empirical analysis of the relationship between demographics and economic growth (GDPGR) is conducted through four distinct variations of the model, with each variation representing one of the four demographic factors of interest in this study. The findings reveal that both population growth (POPGR) and death growth rates (DGR) significantly affect economic growth (GDPGR) in both the short and long-run. In contrast, the birth growth rate (BGR) impacts economic growth (GDPGR) only in the short-run, while net migration growth rate (NMGR) influences economic growth (GDPGR) primarily in the long-run. The study confirms a long-run relationship (cointegration) between demographic variables and economic growth (GDPGR) in the four variations of the model across the provinces in South Africa.
The fourth empirical study employs the autoregressive distributed lag (ARDL) bounds testing technique alongside the Toda and Yamamoto causality test to explore the relationship between demographic dynamics and climate change in South Africa, analysing data from 1995 to 2022. Results from the Bounds F-test confirm a long-run cointegration between carbon dioxide (CO2) emissions and independent variables at a 5% significance level, supported by a significant error correction term (ECT) indicating an 81.75% adjustment speed towards equilibrium after disturbances. While urban population growth (UPGR) and life expectancy (LEXGR) do not significantly influence CO2 emissions in the long-run, they have a significant influence on CO2 emissions in the short-run. Renewable energy consumption (REC) significantly but negatively impacts CO2 emissions in the long-run and has a significant impact
vi
on CO2 in the short-run. The results of the Toda-Yamamoto causality test reveal that LEXGR, and REC Granger-cause CO2 emissions, and conversely, CO2 emissions Granger-cause UPGR, LEXGR, and REC in South Africa. The findings advocate for improved health systems and renewable energy strategies to achieve sustainable resource utilisation and effectively address climate change.