Impact of international trade on employment in orange industry of South Africa

Abstract

The purpose of the study is to analyse the long-run and short-run dynamic relations amongst total employment (lnEMPGt), export output (EXPOt) and import output (IMPOt) from 1990 to 2018, by applying a time-series analysis. The study adopts the secondary data for total employment from the Citrus Growers Association of South Africa, while both export and import output were sourced from the Global Trade Atlas. The multivariate cointegration approach is adopted in the study to identify any causal relationships amongst the concerned variables. The chosen optimum lag selection criterion was the Akaike Information Criterion (AIC) due to its association dependence on the log-likelihood ratio. The third lag was selected for the entire analysis. The results from the cointegration test and the Vector Error Correction Model (VECM) suggest a positive long-run effect between total employment and export output, while import output is negatively associated with total employment. The adjustment term of lnEMPGt, EXPOt and IMPOt suggests that the previous year’s errors are corrected for the current year at a convergence speed of 0.002, 1.11 and 25.37 percentage points, respectively. The results of the Granger causality test show that there are bidirectional causality effects between export output and total employment in the long run, while there are no causality effects between import output and total employment. The overall conclusion is that export outputs positively impact employment, while import outputs impact it negatively in the South African orange industry.

Keywords

• South African orange industry
• employment, wages
• international trade
• Johansen cointegration
• vector error
• correction model