A demand analysis of farm labour employment in the South Coast and Midlands commercial sugarcane farming by labour categories : implications of the sectoral determination.

Abstract

The objective of this study is to investigate the impact of changes to the regulatory environme nt governing farm labour employment since the mid-1990s on wages and employment of farm workers in the South African sugar industry. This study may be differentiated from previous research on the topic, e.g., Conradie (2005), Sparrow et al. (2008), Murray and van Walbeek (2008), Bhorat et al. (2012), and Stanwix (2013) in so far as the impacts on wages and employment are investigated for four categories of farm workers (namely, drivers, seasonal cane harvesting staff, permanently employed elementary farm workers, and casual labour); whereas previous studies have analysed the impacts on aggregate employment. The distinc t ion is important because not all categories of farm workers historically earned wages less than the real minimum wage rate and because the wage elasticities of demand for labour in sugarcane farming are expected to vary by labour category and by sugarcane producing region. Whilst it is well established that the wage elasticity of demand for farm labour in commercial farming in South Africa is relatively price elastic in the long run (Sparrow et al., 2008; Bhorat et al., 2012), no published research has determined the wage elasticities of demand for farm labour in sugarcane farming in South Africa, per se. The time series data on wages and employment of farm workers in sugarcane production by large scale growers (LSGs) from 1978 to 2012 based on Labour Utilisation and Cost Survey (LUCS) were obtained from the South African Cane Growers Association (SACGA). Analysis of the data verifies expectations that the introduction of regulations on wage structure and computation through the Sectoral Determination (SD) for the Farm Worker Sector in 2003 is associated with an increase in real average cash wages and a reduction in the average real absolute value of non-pecuniary benefits (e.g., on-farm accommodation and rations) received by workers in sugarcane farming. Furthermore, the increase in cash-wages outweighed the reduction of non-pecuniary benefits, on average composite wages. Implementation of the SD is also associated with an increase in standardisation of wages for relatively unskilled seasonal and permanent farm employees, which is in line with an intended objective of the SD to standardise the farm wage. Whilst real average wages of workers in sugarcane farming increased by about 70% from 1978 to 2012, employment in sugarcane farming, measured as Full-Time Equivalents (FTEs) Worker, declined by about 36% during the same period, which gives rise to a simple elastic ity computation of -0.51 (Δemployment/Δwage = -0.36/0.7), without accounting for changes in factors other than the wage. In order to account for other factors affecting the supply of and demand for the various categories of farm workers in sugarcane production, econometric techniques were used to estimate the relationships between real farm wages and employme nt in sugarcane farming, and in turn, the wage elasticity of the demand for each category of labour for the industry as a whole and for two particular sugarcane producing regions, the South Coast and Midlands of KwaZulu-Natal (due to the topography of the two regions, sugarcane production on the South Coast is relatively more labour intensive than in the Midlands). High levels of multicollinearity in the data precluded satisfactory estimation of supply and demand functions for farm labour using 2SLS regression techniques. Instead, the Principal Component Analysis (PCA) extraction procedure proposed by Chatterjee and Price (1977) was used to estimate the supply and demand functions separately. Whilst the statistical fit of the estimated farm labour supply (demand) functions was relatively poor and satisfied, respectively, the theoretical fit of the estimated supply of and demand for labour functions was satisfactory. After accounting for changes in the area under sugarcane, the price per ton of sugarcane, price index of chemicals and labour-related policy implications, estimates of the wage elasticities of demand for the various categories of farm labour, regions and aggregate sugarcane production by LSGs ranged between -0.028 and -0.488 in the short-run, and between -0.041 and -0.647 in the long-run. This finding that the demand for farm labour in sugarcane production is relatively inelastic is consistent with observations that, by and large, sugarcane production methods remained relatively unchanged in the industry and each of the two regions for the 1978-2012 period. The results further indicate that farmers adjust employment in response to a wage change within a period of three years. Over-and-above the impact of changes in the policy on wages, the changes in policy are associated with a further reduction in aggregate employme nt of an estimated 4119 FTEs and 5768 FTEs in sugarcane production by LSGs in the short- and long-run, respectively. Other things being equal, considering employment levels in 1978 and 1994, LSGs reduced employment in sugarcane farming by 6.82% and 17.1% in the short- and long-run, respectively. This impact may be ascribed to regulation induced changes in non-wage costs of employment, such as transactions costs and perceived risk. The results further verify that chemicals application is a strong substitute for labour in sugarcane production (especially in the Midlands region), and that employment of farm workers in sugarcane production is positively related to the price of sugarcane and the extent of area planted to sugarcane. Bearing in mind the increases in the real minimum wage of farm workers post 2012, a revision of the current labour legislation to reduce some of the non-wage costs of employing farm workers is recommended to help preserve employment, especially of relatively unskilled workers in sugarcane farming. Activities that increase (reduce) the price received by farmers for sugarcane will have a positive (negative) impact on farm labour employment, both through the price effect, as well as an increase in the area under sugarcane. Finally, considering the trend of declining farm employment, programmes to improve education and training of former farm workers and other people in rural areas are important to improve their prospects of finding employment in non-farm sectors. Furthermore, more research on compliance issues is recommended to improve the effectiveness of compliance on the labour markets that are covered by the SD.