This paper describes formal employment trends in the South African economy since 1970, through using both survey and time-series data. In addition, the study tries to understand the forces that have shaped these employment trends. The descriptive statistics reveal that the primary sectors have shed close to 1.5 million jobs in the period 1970-95. The marginal net employment growth that occurred was to be found primarily in the service sectors. Noticeably, net employment creation in manufacturing was 400 000 jobs over this period. These sectoral trends are matched by occupational trends, which show a significant rise in the demand for highly skilled workers at the expense of unskilled workers. The racial dimension to this is that non- African workers have gained from these labour demand trends, while African workers have lost out significantly.
Utilising an established methodology, the second component of the paper is to try and assess what factors have caused these employment trends which have disproportionately favoured skilled workers. The results of the analysis show that the key cause of the shift toward high-end workers has by and large been technological change within the individual sectors. The rising capital intensity in sectors, coupled with greater computerisation, has prompted the need for more high-end workers. Interestingly though, the results suggest that when examining lower skilled workers, the importance of technological factors remains, but diminishes. Instead, for those at the bottom end, the altering shares in national output of different sectors had a more significant role in understanding relative employment shifts.
The third and final segment of the paper attempts to estimate the impact of trade flows on labour demand, using both survey and time-series data. The results illustrate that the correlation between international trade and employment has been positive. The employment of all workers, by occupation, race and education level, grew as a result of the flows of exports and imports in the economy between 1970 and 1995. However, these gains were not skills, race and education-neutral. Specifically, employment expansion for skilled individuals, or those who had high educational qualifications or workers who were non-African, was appreciably greater than for individuals who did not fit into either of these cohorts. In short, workers at the bottom gained from international trade but gained significantly less than their counterparts at the top end. The time series evidence for manufacturing yields slightly different outcomes. The results here suggest that in the period 1970-1988, unskilled workers gained more than skilled employees did from international trade. However in the subsequent period, and particularly in the 1993-97 period, the reverse effect occurred. In these later years then, there were high job losses for those in unskilled categories, while skilled workers gained significantly.