Wage Discrimination in India's Informal Labor Markets: Exploring the Impact of Caste and Gender

• Published date: 01 February 2013
• Date: 01 February 2013
• Author: Klaus Deininger,Songqing Jin,Hari Nagarajan
• DOI: http://doi.org/10.1111/rode.12020

Wage Discrimination in India’s Informal Labor

Markets: Exploring the Impact of Caste and Gender

Klaus Deininger, Songqing Jin, and Hari Nagarajan*

Abstract

Although there has been considerable interest in wage discrimination in India, available studies have

largely dealt with formal rather than informal markets that are of little relevance for the poorest people.

Focusing on India’s informal labor markets leads to three findings of interest.First, gender wage discrimina-

tion is larger in informal than in formal labor markets, resulting in losses that are larger than receipts from

one of the country’s most important safety-net programs. Second, economic growth will not make gender

discrimination in wage labor markets disappear. Finally, contrary to what is found for gender, the hypoth-

esis of no significant wage discrimination based on caste cannot be rejected.

1. Introduction

Discrimination, defined as a systematic gap in rewards to factors of production that is

due to easily identifiable, though economically irrelevant group characteristics (e.g.

skin color or gender), has long attracted the attention of economists.Even if it implies

sub-optimal outcomes, a caste-based equilibrium can be self-fulfilling and persistent

over time (Akerlof, 1975). While competition can make discrimination disappear

(Becker, 1971), it may persist even in developed countries (Holzer and Neumark,

2000). Moreover, discriminatory practices affect not only current outcomes but also

groups’ social identities and their long-term potential (Hoff and Pandey,2006).

Economic study of discrimination is thus motivated by the moral case for equal

treatment and the potentially far-reaching growth-reducing impacts of discrimination.

The latter arise because discrimination will affect labor supply and lead to underutili-

zation of valuable resources (Klasen, 2002).This can distort incentives for investment,

e.g. in education of girls versus boys based on an expectation that these returns may

be different (Alderman and King, 1998). As a result, inequality traps where a disad-

vantaged group faces a long-run opportunity set much worse than that of another

group even though a better set would be possible (Bourguignon et al., 2007), can arise.

Reinforced by a social order built on endogamous castes—originally based on

occupation—often reinforced by residential segregation, discrimination is particularly

pronounced in India. While caste is losing its relevance for occupational choice in

many urban centers, it continues to exert influence in rural areas of India, especially

Northern ones (Hoff and Pandey, 2006). Caste differences and prejudices may be

* Deininger: The World Bank, MSN MC 3-300, 1818 H St NW, Washington, DC 20433, USA. Tel: (202) 473-

0430; Fax:522-1151; E-mail: kdeininger@worldbank.org. Jin (corresponding author): Michigan State Univer-

sity, Agriculture Hall, 446 W.Circle Dr., Room 211F, East Lansing, MI 48824, USA; Tel: (517) 353-4522; Fax:

432-1800; E-mail: jins@msu.edu. Nagarajan: National Council for Applied Economic Research, Parisila

Bhawan—11 Indraprastha Estate, New Delhi 110002, India; Tel: (91-11) 331-7860; Fax: 332-7164; E-mail:

hknagarajan@ncaer.org. Financial support from the TFESSD Trust Fund and the DFID-World Bank col-

laborative program is gratefully acknowledged. Tore Olsen and Sonal Vats provided excellent research

assistance.The views and opinions expressed in this paper are solely those of the authors.

Review of Development Economics, 17(1), 130–147, 2013

DOI:10.1111/rode.12020

© 2013 Blackwell Publishing Ltd

reinforced by gaps in land and human capital endowments (Deshpande, 2001) or long-

standing bias against females as manifest, for example,in female infanticide (Gleason,

2003). Although the undesirable impacts are well understood,overcoming discrimina-

tion is not easy,partly because efforts at affirmative action often become embroiled in

politics (Deshpande, 2007).

Efforts to quantify the extent of discrimination in India and its evolution over time

will thus be of policy relevance. While some studies have explored statistical discrimi-

nation in wages, they focus on the formal sector only, thus leaving out the roughly

70% of India’s population living in rural areas. This is important because, unless they

have access to resources such as land and capital to allow self-employment, most rural

poor have to rely on casual and informal labor markets to survive. In addition to vola-

tility,e.g. due to climatic shocks, discrimination in these markets will affect the welfare

and scope for investment by the poorest groups in society.

We use a nationally representative household survey to extend analysis of wage dis-

crimination to the informal sector, focusing on three questions, namely to (i) quantify

the size of discrimination in casual labor markets compared to what is found in formal

settings; (ii) explore whether caste- or gender-based discrimination is more important;

and (iii) compare discrimination in villages with high and low income levels. Results

suggest that: (i) discrimination in informal markets is more pronounced than in formal

ones; compared to less than 30% of gender wage gaps in formal labor markets (Jacob,

2006), 61% of gender wage gaps for all casual workers and 76% of the gap for agricul-

tural workers is attributable to discrimination; (ii) wage differences between sched-

uled castes and tribes (SC/STs)1and the rest of the population reflect differences in

endowments rather than discrimination; and (iii) gender discrimination in rich villages

is higher than in poor villages so that growth alone may not make it disappear, as has

been found for dowries (Anderson, 2003).

This implies that ways to improve outcomes for women in casual labor markets may

be justified. Initiatives to expand women’s access to productive assets, e.g. through

inheritance legislation such as recent reforms of the Hindu Succession Act, measures

that make it easier to rent land, or credit access via self-help groups, all go in this

direction. Research to assess the extent to which these measures’ affect discrimination

and female empowerment is desirable.

The paper is structured as follows. Section 2 draws on the literature to set out our

conceptual framework and approach. Section 3 presents data and uses descriptive

evidence to highlight caste- and gender differences in India’s informal sector. Section

4 presents results from wage regressions in agriculture and non-agriculture and

decomposition of wage differentials. Section 5 concludes with a number of policy

implications.

2. Motivation and Framework

The literature points to continued relevance of gender and caste differentials.We thus

discuss techniques to decompose differences in wage rates as well as the assumptions

about the “unbiased” situation underpinning them. This links to a discussion of the

estimation strategy,in particular strategies to control for selection, used in subsequent

sections of the paper.

Discrimination: Empirical Evidence and Policy Implications

Since wage discrimination, defined as the gap in earnings between different types of

workers that remains once all observable characteristics have been accounted for

WAGE DISCRIMINATION IN INDIA’S INFORMAL LABOR MARKETS 131

© 2013 Blackwell Publishing Ltd