CoPS/IMPACT Working Paper Number G-241

Title: Theoretical specification of a labour-supply module, including HIV/AIDS, for South Africa

Author: Louise Roos


This paper describes the application of a detailed labour-supply module, including HIV/AIDS for South Africa. I begin by developing a model of the South African labour market that embodies two salient features of HIV: (1) incidence and transmission rates follow distinct patterns by age, gender, race and occupation; (2) the disease progresses through a number of distinct stages. To model these features, the labour-market model must be dynamic, and identify large amounts of age, gender, race, labour-market function and HIV-stage detail. Dixon and Rimmer (2003) develop and apply a labour-market model with high levels of occupational detail. They subsequently extend this model to investigate issues related to immigration (Dixon & Rimmer, 2009, 2010, 2011), and of relevance to this paper, health (Dixon et al., 2010).

In developing the labour-market module, we define people between the age of 15 and 65 by labour-market function, age, gender, race and HIV stage. We distinguish between employment, unemployment, new entrant and "permanently departed from the labour force" categories and employment, unemployment and "permanently departed from the labour force" activities. People in each category at the start of year decide on their labour supply during year . They make this decision by solving an optimisation problem. We assume that people in employment categories supply their labour more strongly to employment activities than people in the unemployment categories. We also assume that supply of labour depends on health status. A person who is HIV negative supplies labour more strongly to employment activities than a person who is HIV positive. Adults grouped in the "permanently departed from the labour force" category do not offer to any labour force activity. As we shall see, under this modelling framework, policies aimed at reducing the transmission of HIV in year increase labour supply in year by increasing (relative to the basecase) the number of HIV-negative people in year . Although not described in this paper, improved treatment of people already infected with the HIV virus may decrease the rate at which people move through the stages of HIV infection. Improved treatment may increase the effective labour supply of HIV-infected people.

The outline of this paper is as follows. Section 1 defines the key concepts necessary for the functioning of the labour-market specification. Section 2 explains the basic labour-supply mechanism. The aim is to introduce the basic mechanism before it is explained in detail later in this chapter. Section 3 describes how categories at the start of year are determined. Section 4 describes the utility maximisation problem that people face when they decide on their labour offers. This equation shows that labour offers depend on relative wages. Section 5 describes the possible flows between categories and activities. Section 6 describes the equations facilitating these flows. Section 7 explains the wage determination process in the policy simulation. Under this process real wages are sticky in the short term and flexible in the long term, with employment adjusting in the short term and returning to basecase in the long term. Section 8 describes equations that allow for independent HIV forecasts to be introduced to the basecase. Section 9 describes an equation that models the deviation in the number of new HIV cases. This equation can be activated during the policy simulation.

JEL classification: I190, O55.

Keywords: Africa, HIV/AIDS.

Working Paper Number G-241 can be downloaded in PDF format.

Go to working papers page