Skilled-unskilled wage gap without discrimination – the case of India

ABSTRACT

Both skilled and unskilled workers are employed formally and informally in developing economies. However, workers employed informally are often discriminated – they receive lower returns to their individual characteristics compared to those employed formally. What would be the skilled-unskilled wage gap if there were no such discrimination in the labour market? We show that the with-discrimination wage gap is more if the unskilled workers are discriminated more in their employment-share weighted terms. We propose a methodology based on Oaxaca-Blinder decomposition technique to estimate the without-discrimination wage gap, apply it to four rounds of Indian NSSO employment data and find that (i) both types of workers face discrimination, and (ii) the without-discrimination wage gap is significantly less. We also check for the possibility of female workers getting discriminated more in informal employment.

KEYWORDS:

• Skilled-unskilled wage gap
• discrimination
• decomposition
• India

JEL CLASSIFICATION:

• J24
• J31
• O17

Acknowledgments

We are grateful to an anonymous referee for incisive comments that have immensely improved the clarity in the paper.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ See Chamarbagwala (2006), Goel (2017), Kijima (2006), Sahoo and Raa (2012) for India, and Katz and Autor (1999), Acemoglu and Autor (2011) for advanced economies.

² See Bargain and Kwanda (2009), Narayanan (2015), and Tansel and Kan (2012) for evidence.

³ On the definition and measurement of discrimination, see Fortin, Lemieux, and Firpo (2011) and O’Neill and O’Neill (2006).

⁴ Oaxaca-Blinder’s technique – though not without criticism – is widely used to identify and measure discrimination. See for e.g. Bhaumik and Chakrabarty (2018), Blau and Kahn (2016) and Fortin, Lemieux, and Firpo (2011).

⁵ For choice of variables see Goel (2017) and Narayanan (2015).

⁶ Since, from a policy perspective, more the formal employment the better, workers with informal employment are chosen as a reference group.

⁷ Detailed decomposition results are available with the authors.

Additional information

Funding

This work was supported by the IIM Indore.