Obesity-based labour market discrimination in South Africa: a dynamic panel analysis
Apart from obesity-related health care costs in South Africa, obesity is also seen to have far-reaching effects that seep into labour market outcomes. Using National Income Dynamics Survey (NIDS) panel data, this study aims to examine the relationship between body mass index (BMI) and employment status as well as wage levels of individuals to identify the optimal level of BMI from the labour market perspective in South Africa. Thereafter, the article uses ethnicity-backed obesity thresholds to measure the discrimination obese individuals face on the probability of becoming employed and their wages earned once employed.
The econometric analysis uses the OLS probit and tobit regression models as the starting point for analysis. However due to issues of reverse causality, the analysis thereafter utilises a system GMM model to take endogeneity into account. A further Blinder-Oaxaca decomposition technique is used to derive the discrimination component in the system GMM regressions for obese and non-obese individuals. Finallly, gender-specific analysis is undertaken to investigate whether obesity-related discrimination differs between males and females.
The relationship between BMI and employment probability/wages is seen to be non-linear with increases in BMI leading to an increase in the probability of employment and wages up to a threshold beyond which this relationship becomes negative. Based on the system GMM estimation, the optimal BMI for employment probability and wage determination is identified as 30 and 27 respectively. Blinder-Oaxaca estimates show that 90% of the gap in employment status is accounted for by obesity-related discrimination. With regard to wages, obesity leads to a discrimination of 186%. Gen1der-specific Oaxaca analysis found that obese females face discrimination in employment probability of 109% compared to a negative discrimination of −184% for obese males. In determining wages, employed obese females face discrimination of around 73% whereas the discrimination endured by employed obese males is half of this, at 35%.
Our findings reiterate that increasingly obesity has adverse labour market implications. Obesity-based discrimination exists in South Africa and is predominantly faced by obese women entering the workplace and continues in the wage determination of both men and women.
KeywordsObesity Unemployment Wages Discrimination Labour market South Africa
JEL classificationI14 J71 J31
- Arellano M, Bond S (1991) Some tests of specication for panel data:Monte carlo evidence and an application to employment equations. Rev Econ Stud 58(2):277-297Google Scholar
- Arellano M, Bover O (1995) Another look at the instrumental variables estimation of error components models. J Econ 68:29-51Google Scholar
- Asgeirsdottir TL (2011) Obesity & Employment: the case of Iceland. University of Iceland, Department of Economics, ReykjavíkGoogle Scholar
- Blundell R, Bond S (1998) Initial conditions and moment restrictions in dynamic panel data models. J Econ 87:115-143Google Scholar
- Dackleburg M, Gerdtham UG, Nordin M (2014) Productivity or discrimination? An economic analysis of excess-weight penalty in the Swedish labor market. Eur J Health Econ 16(6):589–601Google Scholar
- Department of Health (2015) Strategy for the Prevention & Control of obesity in South Africa 2015–2020. Department of Health, Republic of South AfricaGoogle Scholar
- Fairbrother KA (2009) Healthcare burden of obesity in South Africa: a reflection on the role of government. University of Witwatersrand, Johannesburg Retrieved on 2/03/2017 at: http://wiredspace.wits.ac.za/bitstream/handle/10539/8792/AmyFairbrother_ResearchReport_FinalSubmission.pdf?sequence=1 Google Scholar
- Garcia J, Quintana-Domeque C (2006) Obesity, employment, and wages in Europe. Adv Health Econ Health Serv Res 17Google Scholar
- Lechtenfeld T & Zoch A (2014) Income Convergence in South Africa: Fact or Measurement Error? Stellenbosch Economic Working Papers: 10/14, South AfricaGoogle Scholar
- National Institute for Health and Clinical Excellence (NICE) (2013) Assessing body mass index and waist circumference thresholds ethnic groups in the UK (PH 46) [Online]. Available from: http://www.thehealthwell.info/node/522752 [Accessed: 1st May 2016]
- Pinkston JC (2015) The dynamic effects of obesity on the wages of young workers. University of Louisville, Department of Economics, LouisvilleGoogle Scholar
- Roodman D (2009) How to do xtabond2: an introduction to difference and system GMM in Stata. Stata J 9(1):86–136Google Scholar
- Some M, Rasheed N, Ohonba A (2016) The impact of obesity on employment in South Africa. Stud Econ Econ 40(2):87–104Google Scholar
- Statistics South Africa (2014) Mid-year population estimates 2014. Statistics South Africa, PretoriaGoogle Scholar
- Treasury N (2016) Taxation of sugar sweetened beverages, economics tax analysis chief directorate policy paper. Government of South Africa, PretoriaGoogle Scholar
- Tugendhaft A, Hofman K (2014) Empowering healthy food and beverage choices in the workplace. Occupational Health Southern Africa 20(5)Google Scholar
- Villar JG, Oreffice S & Quintana-Domeque C (2011) Physical Activity and Obesity in Spain: Evidence from the Spanish National Health Survey. The Economics of Sport, Health and Happiness: The Promotion of Well-being Through Sporting ActivitiesGoogle Scholar
- Wittenberg M (2011) The weight of success: the body mass index and economic well-being in South Africa. School of Economics. SALDRU and Data First University of Cape Town, Cape TownGoogle Scholar
- Woolridge JM (2002) Econometric analysis of cross section and panel data. MIT Press, Cambridge (MA)Google Scholar