This paper investigates the aggregate and sectoral public-private remuneration pattern in South Africa from 2001:q1 to 2017:q1. Co-integration analysis confirm a stable, long-run relationship. The adjustment to the deviations from this long-run relationship is strong and significant for public-sector remuneration, while private-sector earnings neither respond to the deviations from the long-run relationship nor lagged changes of public sector remuneration. No individual public-sector remuneration is found to Granger-cause an individual private-sector remuneration. On the other hand, causal relations between private-sector remuneration and public sector remuneration cannot be rejected. A traditional “Dutch-disease” hypothesis for South Africa is rejected. Widening this analysis to individual private and public sectors confirms the results with aggregate earnings with two exceptions: 1) Earnings in financial intermediation and private road transport can be better explained including public sector earnings, and 2) Earnings in manufacturing and mining are found to be related to public sector earnings in the long run. Nevertheless, the degree of fit is low for individual private sector variables except financial intermediation and private road transport while it is high for individual public sector earnings except local authorities. Efforts to slow down the speed of the wage-price spiral should not exclude the private sector.
Aggregate and sectoral public-private remuneration patterns in South Africa
Working Paper 786