Heterogeneity and participation in informal employment among non-cultivator workers in India

Abstract

Labour informality is one of the most serious challenges facing a developing economy like India with large-scale poverty and decent work deficits. This study has inspected possible heterogeneity within informal employment among the non-cultivator workers. Multinomial Logit was applied to find out the determinants of participation in different components of informal employment. Significant heterogeneity within the informal employment on poverty, age, gender, socio-religious communities, educational attainment and industrial classification was observed. There is coexistence of voluntary and involuntary informal employment. Given the diversity of employment, one-size-fits-all policy design may adversely affect sustainable and inclusive growth in India.

Keywords:

• Labour market informality
• heterogeneity
• multinomial logit
• informality as choice
• informality as exclusion

JEL Classification:

• C35
• J16
• J46
• J48

Acknowledgements

We are grateful to Prof. Hiranya Nath, Sam Houston State University, USA;Dona Ghosh, IIT Kharagpur; participants at the ADRI Conference, Patna, India; and to three anonymous referees of this journal for their insightful comments. The usual caveat applies.

Notes

1. The concept of heterogeneity here emphasises the diversity or variation within informal employment. This is in contrast to the homogeneity approach which considers informal employment as a single homogenous group without any dissimilarities or distinctions within it.

2. A stratified multi-stage design is adopted in NSSO EUS where the First Stage Units (FSUs) were the census villages in the rural sector and Urban Frame Survey (UFS) blocks in the urban sector. The ultimate stage units (USU) were households in both the sectors. In all, the number of households (individuals) surveyed in different rounds were- 120,578 (596,686) in 1999–2000; 124,680 (602,833) in 2004–2005; 100,957 (459,784) in 2009–2010 and 101,724 (456,999) in 2011–2012. The entire questionnaire for the NSSO EUS survey is available from their website: http://mail.mospi.gov.in/index.php/catalogue/143/download/1636 for the 68th Round, http://mail.mospi.gov.in/index.php/catalogue/18/download/119 for the 66th Round, http://mail.mospi.gov.in/index.php/catalogue/24/download/176 for the 61st Round, and http://mail.mospi.gov.in/index.php/catalogue/54/download/723 for the 55th Round.

3. NSSO EUS divided enterprises into different sub-types and size-classes as discussed in Appendix 1. Further, workers are grouped into five status codes- own-account self-employed, employer, unpaid family worker, regular salaried, casual wage labour- in public works &other types of works. Additionally, workers were enquired about access to social security benefits such as Provident Fund/Pension, gratuity, health care and maternity benefits etc.

4. The study uses the usual status criteria in determining the employment status. The usual status criteria combine the principal and subsidiary status criteria. NSSO EUS makes a distinction between principal and subsidiary status of a person. The principal status determines the employment status of a person based on the activity on which he has spent a relatively longer time during the last year. The subsidiary status on the other hand is based on the secondary activity (other than the principal activity) of the person performed for a shorter time period not less than 30 days. Under the usual status criterion, hence, workers are those performing any economic activity (work) under either principal or subsidiary status. Thus a person is considered as a worker if he is employed under subsidiary status even if he is not working under the principal status. For further details please refer to National Sample Survey Office (NSSO 2014).

5. The five distinct categories based on ‘usual status’ of the workforce are discussed as follows: Own-account workers are defined as self-employed persons who operated their enterprises on their own-account or with one or a few partners without hiring any labour. Employers on the other hand comprise self-employed persons working on own-account or with one or a few partners and hired labour for their enterprises. Unpaid family workers are self-employed persons who assisted the running of their household enterprise working full or part time without any regular salary or returns for the work performed. Regular workers on the other hand worked in others’ farm or non-farm enterprises, receiving a salary or wages on a regular basis. Finally, casual workers, in contrast to regular employees, are casually engaged in others’ farms or non-farm enterprises and received wages according to the terms of a daily or a periodic work contract.

6. Informal own-account workers and informal unpaid family workers comprise entirely of the own-account workers and unpaid family workers respectively since they do not have access to social security benefits. Informal employers consist of employers in the unorganised sector. Informal regular workers and informal casual workers comprise regular and casual workers respectively without access to social security benefits. Refer to Appendix 1 and Endnote 3 for further details. Since our objective is to investigate the heterogeneity in the characteristics of the informal workers, the study divides the informal workforce on the basis of ‘usual status’ of the workers.

7. Formal employment comprises regular and casual workers with access to social security benefits as well as employers in the organised sector. Refer to Appendix 1 and Endnote 3 for further details. There are broadly two definitions of informality used in the literature. The first is based on the type and size of enterprise leading to organised vs. unorganised sector distinction. The second is based on the access to social security benefits as a criterion leading to the formal vs. informal employment distinction. We use the second of the two definition for our study.

8. To derive our variable, we consider the years of education completed by an individual. For example, a person having completed primary school would have five years of schooling whereas a person with graduate degree will have 15 years of education. For ‘Below Primary’ education, we take the expected years of education i.e. the average of four (Max) and zero (min) years of completed education by the person.

9. Our poverty estimates are based on the Rangarajan Committee methodology (Planning Commission 2014). To arrive at the poverty lines for the earlier years, we deflate the 2011–2012 poverty lines for rural and urban sector separately by the CPI for Rural Workers and CPI for Industrial Workers separately. Our poverty estimates are likely to be different from the Rangarajan estimates due to different data sets and methods employed. Following the methodology proposed by Sengupta, Kannan, and Raveendran (2008), households are classified into five mutually exclusive groups of Poverty Category: (1) Very Poor If MPCE ≤ 0.75 times poverty line (PL); (2) Poor If 0.75 < MPCE ≤ 1 PL; (3) Marginal If 1 PL < MPCE ≤ 1.25 PL; (4) Vulnerable If 1.25 PL < MPCE ≤ 2 PL; and (5) Middle Class and above If MPCE > 2 PL.

10. Of the total population of India in 2001, 80.5% were Hindus while Muslims account for 13.4% and other religious communities 6.1%. In 2011, the figures were: 79.8, 14.2, and 6% respectively. With respect SCs and STs, in 2001 their share was 16.2 and 8.2 percentages respectively; and in 2011 share increased to 16.6 and 8.6%. 2001 population share data were considered as the reference points for 1999–2000 and 2004–2005; and 2011 census data were for 2009–2010 and 2011–2012.

11. We have followed Ministry of Labour and Employment (MoLE) report in classifying workers according to skill levels for National Classification of Occupations 2004 (MoLE n.d.). According to the classification, Legislators, Senior Officials, Managers and Professionals are clubbed together into the Highest Skill jobs. Technicians are grouped into High Skill jobs. Clerks; Service & Sales workers; Skilled Agricultural & Fishery Workers; Craftsmen and Machine Operators are grouped into the Low Skill jobs. Finally, Elementary Unskilled jobs are grouped into Lowest Skill jobs.

12. We have used the concordance tables for NCO 1968 and NCO 2004 to generate the equivalent of NCO 2004 codes for 2004–2005 and classify their skill levels accordingly.

13. The 25 state regions considered in the study include Jammu & Kashmir; Himachal Pradesh; Punjab & Chandigarh; Uttaranchal; Haryana; Delhi; Rajasthan; Uttar Pradesh; Bihar; Tripura; Assam; West Bengal; Jharkhand; Orissa; Chhattisgarh; Madhya Pradesh; Gujarat; Maharashtra, Dadra, Daman & Diu; Andhra Pradesh; Karnataka; Goa; Lakshadweep, Andaman & Nicobar Islands; Kerala; Tamil Nadu & Pondicherry; and North-Eastern states excluding Assam & Tripura.