Confederation of Associations in the Private Employment Sector

The Voice of the South African Staffing Industry


Relative Wage Levels Amidst Legislative Compliance: Remuneration in the Temporary Employment Services Sector in South Africa.


The temporary employment services (TES) sector, colloquially known in South Africa as the ‘labour broker’ sector, has grown rapidly over the last decade. Whilst the TES industry has provided employment opportunities for a significant number of South Africans, the concerns around the industry has been disproportionately related to the wage and non wage benefits received by TES workers. As such, the legislative response has sought to regulate the industry based on concerns around the relative wage levels prevalent in the sector.  This paper seeks to examine wage differentials in greater detail.  
At the aggregate, labour force data suggests that a significant earnings gap exists between TES and non-TES earnings. This was almost 60 percent in 2013 and serves as a significant point of departure for a more nuanced approach to understanding the role of the TES sector in the labour market. However, the nature of the TES industry is heterogeneous and can be divided into two distinct firm types: Firstly, those TES providers many of whom are part of the Confederation of Associations in the Private Employment Sector (CAPES) who are liable for compliance checks and ethical as well as legal obligations enforced by CAPES in terms of wage and non-wage benefits. Secondly, those TES providers who are not registered with CAPES, more commonly known as the ‘bakkie brigade’ where provisions for non-wage benefits are often not included in their wage costs.  
In order to conceptualise an approach that presents the heterogeneity of the industry, we classify workers by the level of compliance of their employer in terms of certain statutory benefits including paid annual leave and paid sick leave and statutory wage costs such as UIF. The compliance categories are as follows:   


  1. Fully compliant (TES statutory wage costs include UIF; TES statutory benefits include paid sick leave and paid annual leave)  
  2. Partially compliant  (Includes one or two of the following but not all three benefits: UIF; paid sick leave, paid annual leave)   
  3. Non-compliant (an employer does not contribute to statutory costs or statutory benefits) 


In 2013, the bulk of TES workers (87 percent) worked in firms that were fully compliant or partially compliant  with statutory costs and benefits outlined in the BCEA. We found that 13 percent of TES workers worked at non - compliant firms and did not have access to any form of non-wage benefits. The initial descriptive evidence presents a significant variation in wage levels within the TES sector, which is driven seemingly by the degree of compliance with the law. The low wages in the TES sector are therefore not necessarily representative of the sector as a whole, but attributable to the non-compliant component of the sector, at least on the basis of the initial descriptive evidence.   The descriptive data also finds that the average lowest wage in non - compliant TES firm is earned by African, males either in the 15 24 or 55-65 age cohort. This is then a proxy for the most disadvantaged TES employee, in the sample of non-compliant firms. The educational differentiation in pay is more pronounced at the top - end of the educational ladder with a university degree or certificate. The differential at the lower end of the education ladder is lowest between fully compliant and non - compliant firms suggesting that earnings for the poorly educated are similar regardless of the level of compliance of firms. In terms of the occupational ladder, ratios of pay differentiation range from two to four times as much when comparing non-compliant and fully compliant firms for the bulk of medium and high-skilled occupations. We find services and sales workers that make up the bulk of the TES sector earn significantly more in the TES sector than outside of it. The wage differential for TES professionals and managers is in part, determined by work experience as well as compliance by the employer. There are certainly cases where given worker experience and a compliant employer, TES and non-TES sector earnings are very similar.  For the sake of accuracy, we pursue a multivariate regression analysis in which we control for the various factors such as firm size, age, gender, sector and experience that may drive wage differentials. The initial results suggest that when controlling for a range of individual characteristics, the wage gap between the TES and non-TES sector increases dramatically as compliance levels decrease. Our OLS earnings function results suggest then, that when controlling for other factors, employees in fully compliant TES firms earn a wage which is on average less than 10 percent of formal non-TES sector wages. This was true for the full sample as well as for Africans. However, when employees are located in non-compliant firms this differential more than quadruples to just under 40 percent relative to the non-TES sector.  This was true for the full sample as well as the sample of Africans and youth. From this, if one were to pinpoint the determinant of low wages in the TES sector then, arguably the key  contributor would appear to be a lack of compliance with the law. 
Whilst the OLS regression is useful, it also examines earnings at the mean instead of at different points along the income distribution. Often these mean results, can mask the returns to TES employment across the entire wage distribution, ranging from low-wage t o high-wage workers. We therefore use a quantile regression to examine wage differentials across the income distribution. An interesting finding presents itself whereby workers in the bottom 40 percent of the wage distribution in fully compliant TES firms, earn more than workers at the lower end of the distribution in non-TES firms. This suggests that low wage workers in fully compliant TES firms earn a premium and may b e less vulnerable to income shocks than those in the non-TES sector, although the premium is small (2 percent). However, for the top 40 percent of the income distribution, on average a worker in the TES sector earns 27 percent less than the non-TES sector for firms that are fully compliant. The earnings differential between non-compliant TES firms is even higher suggesting that the non-compliant TES sector is driving low wages in the sector. Whilst the results for the full sample and African’s are consistent, the results for youth are only consistent with the non-compliant TES sector. 
We conclude that whilst only a minority of TES firms are flouting the law, these firms are driving down wages for the TES sector given the significant differentials. Thus policy responses should revolve around managing compliance levels amongst these firms . This would include dedicating greater resources to inspector and enforcement bodies and creating an incentive to comply with wage legislation utilising an effective penalty structure. In addition, research should be used to present the distinction between fully compliant firms and non - compliant TES firms such that the correct labour brokers are targeted by enforcement bodies. This would ultimately allow the legislative framework that governs the TES sector to focus on those that firms that are non-compliant.  
We also suggest that given the large wage differentials, recent legislative amendments designed to reduce the incidence of TES employment, may in fact be focused on the wrong constraint in the labour market.  If compliance is the predictor of improve d wages in the TES sector, it is not apparent that the recent legislative amendments around the TES sector, have been formulated with these constraints and challenges in the sector in mind.  
Queries addressed to Professor Haroon Bhorat.