Private returns to education for wage-employees and the self-employed in Uganda

Abstract

The paper investigates the differences in private marginal returns to education between wage-employees and the self-employed in Uganda, using the Mincerian framework with pooled regression models. We use a two-wave household panel to estimate homogenous and heterogeneous private returns to education for both worker types. The study finds similar marginal returns to an additional year of schooling for both worker types. Further, the study finds the returns to educational qualifications are convex for wage-employees and concave for the self-employed. With regard to heterogeneous returns to education, we employ quantile regression models and find returns to education decreasing with quantile for both worker types.

Removing the “Veil of Ignorance”: Nonlinearities in Education Effects on Gender Wage Inequalities

“…At the same time, the intense mechanization of several sectors of the Indian economy has decreased the demand for low-skill jobs, especially in casual employment (Fasih et al 2012). This phenomenon is not unique to India and several empirical studies in other developing countries have similarly found that returns to education are highest at the upper-end of the education distribution (e.g., Kavuma et al 2015;Kingdon and Söderbom 2008).…”

“…Thus, a concave relationship between wages or earnings and education is often expected. However, recently, many studies have found that returns to education are highest at the upper-end of the education distribution in developing countries (e.g., Kavuma et al 2015;Kingdon and Söderbom 2008;Söderbom et al 2006;Rankin et al 2010). Some scholars rationalize the convex empirical returns in developing countries as the result of the supply of individuals with low education increasing more than their demand; and reversely the supply of individuals with high education growing slower than their demand (see Fasih et al 2012).…”

Analysing Econometric Bias and Non-linearity in Returns to Education of Pakistan

“…The returns to education estimate may be biased because we take only those individuals who are getting wage or on-job among those who have received an education. The Two Step procedure of Heckman (1979) helps to cope with this problem [Crespo Cuaresma and Raggl (2014); Kavuma, Morrissey, and Upward (2015)]. Wooldridge (2013) stated that if the model description is such in which both endogeneity and sample selection issues exist than in order to simultaneously control for both biases, we should first estimate the selection hazard or Inverse Mills Ratio (IMR) due to non-random sample selection using the Heckman (1979) two-step method and then explicitly include the IMR into the Instrumental Variable (IV) estimation.…”

Returns to education in self-employment in India: A comparison across different selection models

“…A study using models both with and without corrections for sample selection conducted by Rankin et al (2010) for Ghana and Tanzania showed convex returns to education for the self-employed, while another study, by Sandefur et al (2006), for Ghana, Tanzania, and Ethiopia observed convex returns to education for both the self-employed and wage employed in Ghana and Tanzania, but not in Ethiopia. Further, Kavuma et al (2015), using the Mincerian framework with a pooled regression model based on the data for Uganda's labour market, observed a concave relationship between earnings and education for self-employment and a convex relationship for wage employment. Kavuma et al (2015) found that the convex and concave education qualification earnings curve for wage employees and the self-employed, respectively, was primarily reflective of the difference in the demand for skills in the formal and informal sectors, demand being lower in the informal sector.…”

“…Further, Kavuma et al (2015), using the Mincerian framework with a pooled regression model based on the data for Uganda's labour market, observed a concave relationship between earnings and education for self-employment and a convex relationship for wage employment. Kavuma et al (2015) found that the convex and concave education qualification earnings curve for wage employees and the self-employed, respectively, was primarily reflective of the difference in the demand for skills in the formal and informal sectors, demand being lower in the informal sector. Teal (2001), using a model with correction for sample selection bias, noted a similar finding for Ghana, returns to education being far higher for wage earners than for the non-agricultural selfemployed.…”