Wage Gap Between Formal and Informal Regular Workers in India: Evidence from the National Sample Survey

Abstract

This article elucidates the wage differential between formal and informal workers across different sectors, gender, occupation, and industry by using the 61st (2004–2005) and 68th (2011–2012) Rounds of National Sample Survey Office (NSSO) unit-level data. The study emphasizes two things: first, identifying the existence of the absolute wage gap between formal and informal workers and, second, finding the intensity of discrimination in wage between formal and informal workers. The vast body of literature available on this issue identifies gender, caste, religion, and region as the factors causing wage discrimination. This literature makes a shift from these traditional concepts by explaining the importance of job contract as a basis of wage discrimination.

This study utilizes the percentage relative gap (PGR) to work out the absolute wage gap between the two types of workers (formal and informal) and thereafter decomposes it to arrive at the source of the wage gap. The study applies the threefold Blinder–Oaxaca (B–O) decomposition method, which categorizes the total wage gap into three parts. The dependent variable chosen for the equation is the natural logarithm of daily wage. While the wage gap between formal and informal workers did not significantly fall during the study period, the results, on the other hand, indicate that the component of discrimination is larger than the component of endowment. This explains the discrimination perpetrated against informal workers in the Indian labor market. Tackling such discrimination necessitates implementation of more proactive policies for achieving wage equality in India.

Keywords

Wage discrimination formal informal labor market

Introduction

The widespread phenomenon of wage differential in labor markets has notably increased over the past two or three decades for many reasons. Unequal wage is the main source of overall income inequality. In this context, the nondiscriminatory treatment of workers on the basis of their sex, race, and job contracts can be marked as a moral social goal in itself. The annihilation of wage discrimination can also intensify both the efficiency and growth of the overall economy.

The widespread theoretical literature on the economics of labor market discrimination includes studies by Edgeworth (1922), Becker (1957), and Arrow (1973). The empirical analysis of labor market discrimination in a developing country like India has also been undertaken by various sociologists and anthropologists in the past, including Cox (1959), Beteille (1969), Wadhawa (1975), and Berreman (1979). However, these empirical studies basically focused on overall caste discrimination rather than specifically on labor market discrimination.

Other studies in India have analyzed the discrimination of wage using the National Sample Survey Office (NSSO) and the India Human Development Survey (IHDS) data conducted by the National Council of Applied Economic Research (NCAER). Wage discrimination on the basis of gender in India has drawn extensive academic interest since the mid-1990s when gender wage gap became apparent (Bhaumik & Chakrabarty 2008; Deshpande & Deshpande, 1997; Divakaran, 1996; Kingdon, 1997, 1998; Kingdon & Unni, 2001; Sengupta & Das, 2014).

Banerjee and Knight (1985) and Madheswaran and Attewell (2007) used NSSO data to examine caste-based wage differential in the urban labor market. Gaiha et al. (2007) have evaluated wage differential for rural areas of India. Ito (2009) has investigated the same issue for the rural sector in North India on the basis of field survey, while Kijima (2006), Das and Dutta (2007), and Agrawal (2014), have used IHDS data for a pan-India analysis. Deshpande (2001) explored local variations in inter-caste discrepancy using a Caste Deprivation Index (CDI). To understand the discrimination component, Duraisamy and Duraisamy (2005) have estimated the wage equation, using the quintile regression and ordinary least squares (OLS) methods. Attewell and Thorat (2007) studied the overall discrimination in private sector hiring using primary data. All these studies have focused mainly on wage discrimination associated with gender, caste, and region.

Another set of literature asserts that discrimination in wages is driven by productivity difference. The theory of human capital stipulates that the accumulation of human capital through education and training augments the skills, productive capacities, and life-cycle earnings of workers. The association between wage, experience, and education has been well recognized in literature (Becker, 1964; Mincer, 1958, 1974). Some empirical studies in India also suggest that certain disadvantaged social groups face greater difficulties in getting employment as compared to the advantaged social groups because they have to bear higher levels of transaction costs when they enter the labor market (Ito, 2009). Madheswaran and Attewell (2007) point to the large-scale prevalence of pre-labor market discrimination based on gender, caste, and location (both rural and urban). This pre-labor market discrimination poses a considerable barrier in securing regular and better-quality employment. In addition, the human capital theory proclaims that the wage gap persists because of dissimilarities in the level of education, technical skills, and work experience between the two groups of formal and informal workers. Singhari and Madheswaran (2017) analyzed wage differentials in the informal and formal sectors in India, but they have also not incorporated the wage differential for informal and formal workers.

There have been very few studies on wage discrimination based on job contracts. Since the post-economic reforms and rapid structural adjustment, the Indian labor market has witnessed high labor market flexibility and rising informalization of work. Various reports of the Government of India (GoI, 2007, 2012) have pointed to the rapid informalization of work during the past decade. The informal sector plays a key role in the Indian economy, as it provides employment to about 92% of the workforce and contributes almost 50% to the gross domestic product (GDP) of the country (GoI, 2012). The coherence of employment in the formal versus informal sectors was in the proportion 8:92 in 2004–2005, which fell to 7:93 in 2011–2012 (Appendix A).¹ Unfortunately, the formal sector has seen a rising share of informal employment that went up from 46% in 2004–2005 to 58% in 2011–2012, whereas the share of employment in the formal sector declined from 53% in 2004–2005 to 43% in 2011–2012 (see Appendix A). While there has been rapid informalization of the labor force in the overall economy, informalization within the formal sector has also been a notable phenomenon over the past few decades (Gandhi et al., 2013; GoI, 2007; Srija & Shirke, 2014; Srivastava, 2016) see Appendix B. The proportion of informal workers increased by 5% for all workers during the past decade and by10% for regular workers during the corresponding period. A large section of the labor force in the Indian economy comprises informal workers. Wage differential on the basis of job contracts also hampers the development of the overall economy through rising income inequality.

Using the studies listed here, the discussion in the present study focuses on wage discrimination between formal and informal workers. As per our understanding, there has been no or very little work done in this area. In present time, almost 93% of the Indian workforce comprises informal workers whether in formal or informal sector. The focus of the study is to identify the existence and extent of discrimination in the wages between formal and informal workers across different sectors, gender, occupation, and industry by using the 61st (2004–2005) and 68th (2011–2012) rounds of NSSO’s employment–unemployment round unit-level data. Using the threefold Blinder–Oaxaca (B–O) decomposition technique, the findings indicate that the component of discrimination is larger than endowment component. The finding explains job contract as a basis of wage discrimination.

This article is organized as follows: a brief discussion on the background is followed by the theoretical framework of the analysis. The next section delineates the methodology used in the analysis and the data and variables used in this study. The empirical results are then followed by the conclusions and finally the policy implications.

Economic Theories of Discrimination

The theoretical models of discrimination available in the economic literature are classified under two categories: the first is a competitive model in which agents act individually, and the second model is a collective one, where there is collective action of groups against each other (Altonji & Blank, 1999). Fundamentally, the competitive model primarily focuses on two types of discrimination—“taste-based discrimination” and “statistical discrimination.” Becker (1957) establishes the taste-based discrimination in his study. He further accepts that in a competitive framework, there is existence of three separate kinds of discrimination: customer, owner, or employer and the co-worker (employee). Employers may assume that male workers, who constituted the majority group, are more productive than women (minority group) workers, who comprise the minority group.² In case of the first kind of discrimination, the employer is ready to employ workers who are willing to accept a wage less than what is paid to their counterparts for being either equally productive or more productive at that particular wage. Further, employee discrimination is said to exist when dissimilar employees prefer not to work with colleagues who belong to a different gender, caste, and/or race. Consumer discrimination exists when consumers are not willing to buy goods and services from associates of a certain group and as an alternative are ready to pay a higher price.

The literature on the statistical theory of discrimination consists of two aspects (Altonji & Blank, 1999). While the first depicts that the hiring and wage decision by an employer is influenced by the prior beliefs regarding the productivity of the members in the group (Arrow, 1973), the second issue is related to the repercussion that arises due to group differences, based on the knowledge acquired by the employers, regarding productivity of individuals, and this has been addressed by Aigner and Cain (1977). There have been superior statistical discrimination models (Aigner & Cain, 1977; Arrow, 1973; Phelps, 1972), which argue that women (and those who are in minority groups) receive low wages because, on an average, these groups exhibit low productivity and skill component. The statistical model assumes that the firms have negligible or less information about the productivity and skills of the employees, and the employer’s judgments are based on the basis of incomplete information. Thus, employers use characteristics like gender or social group as a proxy to judge their productivity and skill. In other words, if employers consider that women (or disadvantaged social groups) are normally less productive than men (seen as an advantaged social group), they use the gender or race to which the individual belongs as a screening tool while hiring or making salary or wage decisions. This study follows Becker’s test-based discrimination to estimate wage discrimination between formal and informal workers.

Data and Materials

This section provides definitions of the formal and informal sectors, as well as of other concepts, such as employment, methodology, data, variables, and deflators used in this analysis.

Formal–Informal Sectors and Employment

There is a huge debate on the formal and informal sectors, as well as employment, in the existing literature. Prior to International Conference of Labor (ICLS) 2003, the classification between formal and informal was based on enterprise. Later on, the notion of informality shifted from enterprise to employment. The Government of India constituted a National Commission on Enterprises in the Unorganized Sector (NCEUS) to identify and analyze different dimensions of the informal economy. This study follows the definition of informal³ employment as proposed by NCEUS.

As per NCEUS (2007) “Unorganized workers consist of those working in the unorganized enterprises or households, excluding regular workers with social security benefits, and the workers in the formal sector without any employment/social security benefits provided by the employers.”

Assuming the existence of a dichotomy between formal and informal workers in the Indian labor market, which would create a difference in wage determination, this article tries to analyze the wage structure; wage differentials between formal workers and informal workers by gender; the rural–urban dividend among workers, industry, and occupations; and general and technical educational status.

Methodology

This study applies the PRG to observe the wage differential in absolute terms between formal and informal workers, and then decomposes wage to identify the basis of the gap. The study identifies three dissimilar methods to estimate wage discrimination in the empirical literature. The first is a single-equation technique that treats “informal” and “formal” workers as predictors, while predicting their earnings based on the characteristic of all the works performed by them (Mincer, 1974). The single-equation technique allows for wage structure to be similar for formal and informal groups, thereby providing a reason for the biased estimate. Further, this method limits the value of the coefficients for other independent variables, like experience and education among others, considering them as equal for both the formal and informal wage workers (Gunderson, 1989; Madheswaran & Attewell, 2007).⁵ The second method pertains to the decomposition method, whereby the real wage differential between an “endowment” component (identifies the contribution of differences in the explanatory variables across the groups) and the “coefficient” (measures the difference in groups) component is separated. The final result provides a “discrimination” coefficient, which arises out of unexplained part. This methodology was first evolved by Blinder (1973) and Oaxaca (1973) and used as one of the most comprehensive tools to incorporate the “selectivity bias” (Reimer, 1983) and the “index number problem” analyzed by Cotton (1988) and Neumark (1988). The last method proposed by Brown et al. (1980) refers to an “expanded approach,” which includes the allocation of work into the earnings estimation. An improvement associated with this method is that wage discrimination and job discrimination may be estimated simultaneously. The present study applies the decomposition technique to find the discrimination between the wages earned by formal and informal workers in the same occupation and industry with the same set of endowments, but with the existence of wage discrimination because they have different job contracts. In this section, we have used the symbols “f” for formal worker and “if” for informal worker.

The Blinder–Oaxaca Decomposition

The B–O decomposition permits the classification of wage gap into two parts: one which explains wage gap due to difference in the individual characteristics and the other wage gap, which cannot be explained by the differences in characteristics of the individuals, that is, what may be termed as “discrimination.” The gross wage gap between formal and informal wage workers can be defined in terms of the following equation:

G = \frac{W_{f} - W_{i f}}{W_{i f}} = \frac{W_{f}}{W_{i f}} - 1

(1)

where there are two groups: the first is a formal worker denoted by “f” (advantaged) and the second is an informal worker denoted by if (disadvantaged). In the absence of wage discrimination in the labor market, f and if wage differentials reflect differences in pure productivity:

Q = \frac{W_{f}^{0}}{W_{i f}^{0}} - 1

(2)

where the superscript zero represents the absence of wage discrimination between formal and informal workers in the labor market. The wage discrimination in the labor market coefficient (D) is then defined as the proportionate difference between G + 1 and Q + 1, as follows:

D = \frac{(\frac{W_{f}}{W_{i f}}) - (\frac{W_{f}^{0}}{W_{i f}^{0}})}{(\frac{W_{f}^{0}}{W_{i f}^{0}})}

(3)

Equations (1)–(3) implies:

ln (G + 1) = ln (D + 1) + ln (Q + 1)

(4)

which represents logarithmic decomposition of the gross wage differential.

This OLS equation is written as:

ln {\bar{W}}_{f} = \sum {\hat{β}}_{f} {\bar{X}}_{f} + ξ_{f} (formal workers’ wage equation)

(5)

ln {\bar{W}}_{i f} = \sum {\hat{β}}_{i f} {\bar{X}}_{i f} + ξ_{i f} (informal workers wage equation)

(6)

where $\ln \bar{W}$ denotes the geometric mean of earnings, $\bar{X}$ the vector of mean values of the regressor, $\hat{β}$ the vector of coefficients, and ξ is the error term. The gross difference in logarithmic terms is as follows:

\begin{array}{l} ln (G + 1) = ln ({\bar{W}}_{f} / {\bar{W}}_{i f}) = \ln {\bar{W}}_{f} - \ln {\bar{W}}_{i f} \\ = \sum {\hat{β}}_{f} {\bar{X}}_{f} - \sum {\hat{β}}_{i f} {\bar{X}}_{i f} \end{array}

(7)

If, for a given endowment, due to absence of discrimination, the informal workers are paid as per the formal worker wage structure, then the informal worker wage function is given as:

\ln {\bar{W}}_{i f} = \sum {\hat{β}}_{f} {\bar{X}}_{i f}

(8)

Subtracting Equation (8) from Equation (7), we get:

\ln {\bar{W}}_{f} - \ln {\bar{W}}_{i f} = {\sum \hat{β}}_{f} ({\bar{X}}_{f} - {\bar{X}}_{i f}) + \sum {\bar{X}}_{i f} ({\hat{β}}_{f} + {\hat{β}}_{i f})

(IX)

The B–O decomposition could face the “index number problem,” that is, the choice of wage structure which is nondiscriminatory. The B–O approach assumed that in absence of discrimination, the formal workers’ wage structure would apply for both formal and informal workers. Some researchers take the mean of the estimates found from the two equations to check out the index number problem (Greenhalgh, 1980). Therefore, the wage structure of both groups has been used in order to recognize the sensitivity of the results. The decomposition equation by using reference wage structure of the informal worker (as second group) can be written as:

\ln {\bar{W}}_{f} - \ln {\bar{W}}_{i f} = {\sum \hat{β}}_{i f} ({\bar{X}}_{f} - {\bar{X}}_{i f}) + \sum {\bar{X}}_{f} ({\hat{β}}_{f} + {\hat{β}}_{i f})

(10)

The first part of the right-hand side of Equation (10) explains the difference in endowments like education and experience, among others. This implies the difference in wages, which may be explained by the mean differences in the productivity characteristics of formal and informal wage workers. The second term of this equation expresses the components pertaining to discrimination. The unexplained component in the equation depicts the difference between returns to identical attributes of the formal and informal workers.

Data and Variables

In this study, we use two rounds of the national representative, cross-sectional survey conducted by the NSSO, under the Ministry of Statistics and Program Implementation (MoSPI). The NSSO conducted employment–unemployment surveys during the period between 2004–2005 (the 61st Round) and 2011–2012 (the 68th Round). The study has used these two rounds of NSSO unit-level data for the empirical analysis because of two reasons: first, it provides all information regarding work status and job contract of the worker that helps us to estimate informal and formal wage workers. Second, it is a recently published national representative data on the Indian labor market. Generally, these data are used to estimate occupational distribution, labor force participation rate and wage rates. The method used to collect the sample is two-staged stratified random sampling. In the first stage, village and urban blocks are considered as a unit. The second stage selects households as a unit from the selected urban blocks and villages. The survey details and the aggregate estimates have been delineated in the NSSO reports.

Dependent Variables

The dependent variable taken for the wage function is the natural logarithm of daily wage for a regular worker. The sample used is controlled to persons who work as a regular wage worker, aged 15–65 years. To account for robustness, the wage distribution was trimmed by 0.1% at the top and bottom tails. The nominal wages were converted to 2011–2012 prices using the consumer price index for industrial workers (CPI-IW) for urban workers and the consumer price index for agricultural labor (CPI-AL) for rural workers (RBI, various years). The employment and unemployment surveys in India do not provide any wage data on an hourly basis. Based on the current daily status (CDS), the NSSO provides information regarding weekly salaries of the labors. The daily wages were calculated by dividing the sum of the salary received by workers within the reference week by the aggregate number of days of work in the corresponding reference week.

Control Variable

The educational levels of an individual are categorized under the following six categories: illiterate (below primary), primary, middle, secondary, higher secondary, and graduate. In this study, the general education level code given by NSSO has been converted into the year of schooling (see Appendix C). Potential experience, which is calculated as the difference between age and the number of years of schooling plus six, has been considered as a proxy for the actual experience (Mincer, 1974). It has been assumed that schooling starts at the age of 6 years, and an individual starts working right away after schooling. We also control technical education and vocational training as the human capital variables.

The control variables (demography) used in the form of dummy variables are gender, place of residence (urban/rural location), membership of a social group, religion, marital status, and region (region dummies to capture regional variations).The scheduled castes (SCs) and scheduled tribes (STs) are two historically disadvantaged groups in India, and while examining wage differences on the basis of job contracts, the present study regroups them into two groups: upper castes [including the Other Backward Castes (OBCs) and Others], and lower castes (SCs/STs).

Results

Evidence from Wage Differentials

A considerable difference has been found between the average daily wages of formal and informal workers. This study examines the formal–informal wage differential across sectors and gender in terms of the PRG.⁵ The PRG, in the average daily wage earnings, reflects the rational measurement of the difference between the daily wage earned by a formal worker and that earned by an informal worker. In brief, the PRG measures the proportionate difference in daily wages of formal and informal wage workers. Table 1 shows a decline in the PRG between formal and informal workers, for both males and females, during the study period. The overall PRG for all individuals decreased from 69% to 61%. This decrease in PRG also implies a narrowing down of the wage gap between formal and informal wage workers during the study periods. Table 1 shows the comparative PRG by sector, that is, rural or urban over the period of time. After the rapid informalization, the rural sector has been witnessing the process of breaking of the rigidity of structure of the wage market. The PRG of the rural sector fell from 64% to 55%, whereas the urban sector saw a comparatively lesser reduction rate of PRG from 67% to 60%. The relatively sharp decline in the PRG of the rural sector is due to the lower number of working days for informal workers. On the same lines, the rising demand for labor in rural areas and rural to urban migration (leading to shortage of labor in rural areas) is responsible for a rise in the wages of a contractual worker. In contrast, the urban sector did not match the improved wages of informal workers and their huge supply. Therefore, the reduction in the PRG of the rural areas was higher than that in the urban areas.

Table 1.

Average Real Daily Wage (in ₹). Rate for Formal and Informal Workers by Sector and Gender.

	2004–2005			2011–2012
	Formal Workers	Informal Workers	PRG	Formal Workers	Informal Workers	PRG
Male	547	180	67	729	294	60
Female	466	109	77	634	204	68
Rural	405	145	64	492	221	55
Urban	573	190	67	774	307	60
India	538	168	69	716	277	61

Source: Computed by author from various rounds of NSSO unit-level Employment–Unemployment data.

The average wage rates, the difference therein, and the PRG between formal and informal workers for different industries reflect to a large extent differences in the existing wages; informal workers were paid low wages more often than formal workers. This indicated a significant difference in the existing wages between formal and informal workers within the same type of economic activity. Table 2 shows that irrespective of the affiliation of the workers from any industry, the earnings of informal workers are less than that of formal workers. The wage differentials are highest for public administration and other services, followed by transport, storage and communication, mining and quarrying, and manufacturing (agro-based) during the study period. The earnings of informal workers were approximately at par with formal workers only for the manufacturing (other) sector as the PRG for this sector dwindled from 53% in 2004–2005 to 36% in 2011–2012, implying a reduction in wage differential by almost 20 points. However, the PRG was found to be fluctuating in the transport, storage and communication, manufacturing (agro-based) and trade, and hotel and restaurant sectors by approximately 15–20 points during the study period. For the remaining sectors, though the change during the study period was negligible, the wage differences and PRG were substantial and significant over the period of the study. This shows that a significant wage gap exists between formal and informal wage workers within the same industry, and also that the wages for formal workers were more often higher than those for informal workers.

Table 2.

Average Real Daily Wage (in ₹) Rate for Different Formal and Informal Workers by Industry.

	2004–2005			2011–2012
	Formal Workers	Informal Workers	PRG	Formal Workers	Informal Workers	PRG
Agriculture and allied activity	442	177	60	538	188	65
Mining and quarrying	285	110	61	356	107	70
Manufacturing (agro-based)	252	128	49	321	97	70
Manufacturing (other)	196	92	53	256	165	36
Electricity, water supply, etc.	115	50	57	138	67	51
Construction	238	104	56	287	118	59
Trade, hotel and restaurant	208	130	37	364	173	52
Transport storage and communication	419	171	59	673	176	74
Financial, business and professional services	535	218	59	689	273	60
Public administration and other services	786	135	83	987	199	80

Source: Computed by author from various rounds of NSSO unit-level Employment–Unemployment data.

Table 3 reports the occupation-wise formal–informal wage differentials. The wage differentials are highest for administration and sales sectors. The PRG was found to be fluctuating for administrative, teaching, nursing, and midwifery professionals by approximately 20 points during the study period. In general, the PRG between formal and informal workers within the same types of occupation, for all the categories, was substantial and significant, indicating significant difference in the existing wages between formal and informal workers in the form of higher wages for formal workers.

Table 3.

Average Real Daily Wage (in ₹) Rate for Formal and Informal Workers by Occupation.

	2004–2005			2011–2012
	Formal Workers	Informal Workers	PRG	Formal Workers	Informal Workers	PRG
Administrative	785	468	40	1,282	512	60
Teaching, nursing, and midwifery Professionals	476	390	18	1,024	622	39
Professionals other than teaching, nursing, etc.	373	156	58	708	413	42
Clerical	259	148	43	565	343	39
Service	278	82	70	471	238	50
Sales	177	84	52	537	194	64
Agriculture and allied occupations	178	64	64	271	170	37
Production	226	91	60	480	222	54
Elementary	161	56	65	333	153	54

Source: Computed by author from various rounds of NSSO unit-level Employment–Unemployment data.

Table 4 summarizes the PRG between formal and informal workers on the basis of general educational status. There exists a significant difference between these two types of workers at all levels of education during the study period. For primary education, the PRG was around 60% for 2004–2005 and 66% for 2011–2012 rounds. For different levels of education such as middle and secondary, higher secondary, and diploma, though the PRG figures declined, they were still significant at 55%, 41%, and 34%, respectively, in the 2011–2012 round. The PRG figures for the diploma, graduate, and postgraduate and above levels recorded PRG figures of 44%, 41%, and 54%, respectively in 2004–2005, and 34%, 48%, and 40%, respectively, in 2011–2012. There was a considerable wage differential between formal and informal workers with the same level of education over the study period. Informal workers earned lower wages as compared to formal workers even though they had the same level of education during the study period.

Table 4.

Average Real Daily Wage (in ₹) Rate of Formal and Informal Workers by General Educational Status.

	2004–2005			2011–2012
	Formal Workers	Informal Workers	PRG	Formal Workers	Informal Workers	PRG
Up to primary	309	123	60	618	210	66
Middle and secondary	385	167	57	546	243	55
Higher secondary	515	224	57	611	363	41
Diploma	550	309	44	566	376	34
Graduate	676	397	41	923	477	48
Post-graduate and above	1,086	502	54	1,178	711	40

Source: Computed by author from various rounds of NSSO unit-level Employment–Unemployment data.

Table 5 summarizes the PRG between formal and informal workers on the basis of technical education. There exists an enormous difference between the wages of formal and informal workers at all levels of technical education during the study period. The average wage of workers with no technical education increased for both formal workers and informal workers, but the PRG remained unchanged in both the rounds. For workers with technical degrees and diplomas, and graduates and above, the PRG increased from 17% and 41% in 2004–2005 to 22% and 56% in 2011–2012, respectively. The average wage of workers with diplomas and degrees below the graduate level increased for both formal and informal workers, but the PRG figures remained almost unchanged, at 45% in 2004–2005 and 44% in 2011–2012. This clearly shows a worsening of the situation with regard to the existing wage differential between formal and informal workers even when they had the same levels of technical education.

Madheswaran and Attewell (2007) found substantial discrimination related to employment, but the discrimination occurred to a great extent in the form of unequal access to occupation rather than wage discrimination. As per the above-mentioned findings, there exists wage discrimination between formal and informal workers at each level of general and technical education. The existing discrimination accounts for the wage difference, implying the existence of wage discrimination between formal and informal workers with the same levels of education and skills, and the absence of any pre-labor market discrimination. The wage discrimination implying the disbursal of unequal salaries within a given occupation, despite the same levels of education and skills among both types of workers, is more significant than the prevalence of unequal access in certain occupations due to the existence of pre-labor market discrimination in the Indian labor market.

Table 5.

Average real Daily Wage (in ₹) Rate of Formal and Informal Workers by Technical Educational Status.

	2004–2005			2011–2012
	Formal Workers	Informal Workers	PRG	Formal Workers	Informal Workers	PGR
No technical education	451	178	61	677	265	61
Technical degree	1,331	1,111	17	1,105	860	22
Diploma—below graduate	565	312	45	697	390	44
Diploma—above graduate	721	425	41	1,203	534	56

Source: Computed by author from various rounds of NSSO unit-level Employment–Unemployment data.

Decomposition Results

Table 6.

Blinder–Oaxaca Decomposition Results.

	Person		Male		Female
	2004–2005	2010–2011	2004–2005	2010–2011	2004–2005	2010–2011
Formal wage workers	6.08	6.25	6.27	6.09	5.97	6.14
Informal wage workers	5.08	5.27	5.37	5.21	4.38	4.83
Raw difference	0.99	0.98	0.89	0.87	1.59	1.31
Adjusted difference	0.55	0.69	0.7	0.56	0.91	0.97
Percentage explained by endowment	44.4	29.48	21.82	35.5	42.53	25.32
Discrimination as a percentage of the Total	55.5	70.52	78.17	64.49	57.46	74.67

Source: Computed by author from various rounds of NSSO unit-level Employment–Unemployment data.

Notes. (a) A positive number indicates an advantage to formal wage workers; (b) Negative numbers indicate an advantage to informal wage workers.

In this study, we have estimated distinct wage equation for formal and informal workers over the study period and then decompose it for all workers and sectors (formal and informal). In this sample, the average of log wage was 6.08 for formal workers and 5.08 for its counterpart, which yielded a wage differential of 0.99 in 2004–2005, falling marginally to 0.98 in 2011–2012 (Table 6). The wage differential between formal and informal workers reduced marginally during the study period. The study has used the threefold B–O decomposition method that segregates the total wage gap into three parts. The first represents the endowment, which indicates a mean increase in the wages of informal workers having similar characteristics as formal workers. The second is the coefficient, which means change in the wages of informal worker when the coefficient of formal workers is applied to the informal workers’ characteristics. The third part computes the simultaneous impact of endowment and coefficient—called interaction.

The short-run variation in labor supply or demand could be due to any reason, but the difference in the quality of labor or the quality of the job is a major explanation for the wage differential between formal and informal workers in the ordinary labor market. The efficiency wage models (Katz, 1986; Stiglitz, 1986) and geographical factors can explain some of this variance in wages. In general, diverse factors account for the dissimilarity in wages related to workers who perform similar types of work. Further, there are multiple economic, social, cultural, and demographic components that encroach upon labor supply and thus preserve the disparities in wages across sectors and contracts. The quality of schooling and work experience also matter even if the level of education may be the same. Nevertheless, wage differences are seen to be significant even after the upgradation in skills from the secretarial level. Table 5 highlights that the component of discrimination is more than endowment over the period of the study across sectors (both rural and urban) and gender (both male and female). Nevertheless, endowment accounts for 44% (2004–2005) and 29% (2011–2012) of low wages of informal workers when compared to the wages of formal workers in the Indian labor market. This implies that the wage differential because of difference in productivity decreased over the study period. Kumar and Mishra (2008) also point out that the large difference in wages across Indian industries is not explained by the difference in skills. Liberalization promotes the use of capital-intensive technology in developing countries, and this technology needs a highly skilled labor force (Pack & Todaro, 1969). However, the labor markets in developing countries do not support the use of advanced technology because of either the lack of availability of skilled labor or the high costs of employing such labor. During the consolidated era of economic reforms, Indian industry faced highly volatile competition, and firms adopted a hire and fire policy to deal with this competition (Tendulkar, 2004). Our data also showed a sharp increase in the number of highly educated (skilled) workers working as informal workers during the study period. This shows the informalization of skilled jobs in India after the introduction of economic reforms. The informalization of skilled jobs also suggests the complete or significant absence of pre-labor market discrimination for informal workers.

Table 6 shows wage gaps of 55% and 70% between formal and informal workers, stemming only from discrimination, during 2004–2005 and 2011–2012, respectively. The decomposition result shows that discrimination increased during the study period for all types of workers. Two points help explain this discrimination in wages in the Indian labor market on the basis of formal and informal job contracts. First, the informal labor market was treated as a secondary labor market in developing countries (Anderson, 1987). Papola (2013) argues that the Indian labor market has also been segmented on the basis of informal and formal workers. The segmented labor market theory is based on institutional factors, which basically support the demand-side explanation, and rejects the supply-side explanation of the labor market as proposed in neoclassical theory. Hence, wage discrimination arises between informal and formal workers because of the characteristics of the job contract rather than a difference in employee attributes such as education (general and technical), training (vocational and on the job), and experience.

The above-mentioned discussion indicates that informal workers perform the same tasks as the formal workers but at lower wages. Institutional barriers to labor mobility in developing countries (Anderson, 1987) and limited intra-industry labor mobility in India (Mitra, 2016) perhaps could explain this. If labor is authorized to shift without restraint across markets (primary to secondary and vice versa), there would be no reason for such wage variations to exist for a given type of labor.

The structural adjustment program encourages flexibility in the labor market, which, in turn, motivates employers to hire informal workers at a low cost. Contractualization leads to wage variations across Indian industries (Mitra, 2016). The bargaining model (Dickens, 1986) of wage determination shows that workers acting collectively have greater bargaining power than workers acting individually. Unfortunately, however, informal workers are not unionized because they do not have a long-term written job contract, which prevents them from bargaining with their employers.

Conclusion

Wage structures have been segmented by employer attributes rather than by worker merits. Since, informal workers, in both the informal and formal sectors, do not receive the same remunerations that the directly recruited regular workers receive, the consequence is an alarming rise in recruitment of contractual workers—popularly called the “contractualization process,” which has led to a significant reduction in wages. The outcome of outsourcing of manufacturing as well as services from the regulated economies to free economies that practice contractualization is a decline in both wages and working conditions. The decomposition results show a wage differential between formal and informal wage workers, and they come from both endowment and discrimination. But wage differential due to discrimination was higher as compared with endowment during the study period for both genders. On the other hand, the discrimination component has been stronger over a period of time. Hence, on the basis of findings of this study, we can argue that there could be pre-labor market discrimination or inequality in opportunity among wage workers. But the wage differentials between these two groups are largely explained by discrimination on the basis of job contract rather than inequality in opportunity or pre-labor market discrimination. So, we can say wage differentials between informal and formal wage workers are mainly because of discrimination, but endowments also play a significant role in this differential.

Appendix A. Percentage Distribution of Worker by Sector and Employment Type.

Employment Type

Informal Formal

Sector 2004–2005

Unorganized 99.6 0.40

Organized 46.6 53.4

Total 92.4 7.60

2011–2012

Unorganized 99.5 0.50

Organized 58.3 43.4

Total 92.9 6.90

Source: Computed by author from various rounds of NSSO unit-level Employment–Unemployment data.

Appendix B. Percentage of Informal Workers (without contract) in the Formal Sector.

Formal Sector—NCEUS Definition Formal Sector—Alternative (official) Definition

Regular Workers Regular Workers

2004–2005 2011–2012 2004–2005 2011–2012

Primary 26.9 27.6 23.4 22.2

Secondary 60.2 69.7 58.2 68.9

Tertiary 31.7 41.2 26.6 35.6

Non-agriculture 40.3 50.3 36.1 46.4

Total 40.3 50.2 36.1 46.7

Source: Srivastava (2016).

Appendix C.
Descriptive Statistics of Main Variables in the Earnings Function (2011–2012).

Description of the Variables Formal Workers Informal Workers

Variables Mean SD Mean SD

lndwage Natural Log of daily wage (₹) 6.06 0.89 5.22 0.81

age Age in years 38.10 11.13 33.96 11.74

Age Age square (in years) 1575.42 883.80 1291.10 893.02

Pexp Potential experience of the worker 21.71 12.45 20.75 12.80

dBP If the worker has below primary level of education = 1; 0 otherwise 0.16 0.37 0.32 0.47

dPrim If the worker has primary level of education = 1; 0 otherwise 0.09 0.28 0.13 0.34

dMid If the worker has middle level of education = 1; 0 otherwise 0.09 0.29 0.17 0.37

dSec If the worker has secondary level of education = 1; 0 otherwise 0.14 0.34 0.13 0.34

dhsec If the worker has higher secondary level of education = 1; 0 otherwise 0.12 0.32 0.09 0.29

ddedu If the worker has diploma degree = 1; 0 otherwise 0.04 0.19 0.03 0.18

dgaa If the worker has a graduate and above degree = 1; 0 otherwise 0.24 0.43 0.09 0.29

Technical education If the worker has a technical degree = 1; 0 otherwise 0.14 0.34 0.06 0.24

Voc. training If the worker has any vocational training = 1; 0 otherwise 0.35 0.48 0.39 0.49

Male If the worker’s sex is male = 1; 0 otherwise 0.56 0.50 0.61 0.49

Urban If the worker’s place of residence is urban = 1; 0 otherwise 0.69 0.46 0.67 0.47

Caste If the worker belongs to SC or ST = 1; 0 otherwise 0.81 0.39 0.79 0.41

Married If the worker is married = 1; 0 otherwise 0.55 0.50 0.50 0.50

Muslim If the worker follows the Islam religion = 1; 0 otherwise 0.05 0.21 0.12 0.33

Other If the worker follows any other religion = 1; 0 otherwise 0.07 0.26 0.06 0.24

North If the worker is working in the north zone = 1; 0 otherwise 0.15 0.36 0.14 0.35

Central If the worker is working in the central zone = 1; 0 otherwise 0.14 0.35 0.10 0.30

East If the worker is working in the east zone = 1; 0 otherwise 0.16 0.37 0.11 0.31

West If the worker is working in the west zone = 1; 0 otherwise 0.28 0.45 0.26 0.44

South If the worker is working in south zone = 1; 0 otherwise 0.25 0.43 0.37 0.48

Agriculture If the worker is working in agriculture = 1; 0 otherwise 0.02 0.15 0.02 0.14

Manufacturing If the worker is working in manufacturing = 1; 0 otherwise 0.18 0.38 0.32 0.47

Services If the worker is working in services = 1; 0 otherwise 0.80 0.40 0.66 0.47

Source: Computed by author from various rounds of NSSO unit-level Employment–Unemployment data.

Footnotes

Acknowledgments

Author acknowledges N. K. Mishra (Professor, Department of Economics, BHU, Varanasi) and P. P. Sahu (Head of Department, Center for Entrepreneurship Development, NIRDPR, Hyderabad) for important comments.

Declaration of Conflicting Interests

Funding

The author has not received any kinds of financial support for the research, authorship, and/or publication of this article.

Notes

References

Agrawal

(2014). Gender and caste-based wage discrimination in India: Some recent evidence. Journal for Labor Market Research , 47(40), 329–340.

Aigner

D. J.

Cain

G. G.

(1977). Statistical theories of discrimination in labor markets. Industrial and Labour Relations Review , 30(2), 175–187.

Altonji

J. G.

Blank

R. M.

(1999). Race and gender in the labor market. In Ashenfelter

& Card

(Eds.), Handbook of labor economics (pp. 3143–3259). Elsevier.

Anderson

P. Y.

(1987). Informal sector or secondary labour market? Towards a synthesis. Social and Economic Studies , 36(3), 149–175.

Arrow

K. J.

(1973). The theory of discrimination. In Ashenfelter

& Rees

(Eds.), Discrimination in labour markets . Princeton University Press.

Attewell

Thorat

(2007). The legacy of social exclusion. Economic and Political Weekly , 42(41), 4141–4145.

Banerjee

Knight

J. B.

(1985). Caste discrimination in the Indian urban labor market. Journal of Development Economics , 17(3), 277–307.

Becker

G. S.

(1957). The economics of discrimination . University of Chicago Press.

Becker

G. S.

(1964). Human capital—A theoretical and empirical analysis with special reference to education . Columbia University Press.

10.

Berreman

G. D.

(1979). Caste and other inequities . Folklore Institute.

11.

Berreman

G. D.

(1972). Social categories and social interaction in urban India. American Anthropologist, 74(3), 567–586.

12.

Beteille

(1969). Castes old and new essays in social structure and social stratification . Asia Publishing House.

13.

Bhaumik

S. K.

Chakrabarty

(2008). Does move to market have an impact on earnings gap across gender? Some evidence from India. Applied Economics Literature , 15(8), 601–605.

14.

Blinder

A. S.

(1973). Wage discrimination: Reduced form and structural estimates. Journal of Human Resources , 8(4), 436–455.

15.

Brown

R. S.

Moon

Zoloth

B. S.

(1980). Incorporating occupational attainment in studies of male/female earnings differentials. Journal of Human Resources , 15(1), 3–28.

16.

Cotton

C. J.

(1988). On the decomposition of wage differentials. Review of Economics and Statistics , 70(2), 236–243.

17.

Cox

O. C.

(1959). Caste, race and class. A study in social dynamics . Monthly Review Press.

18.

Das

M. B.

Dutta

P. V.

(2007). Caste matter for wages in the Indian labor market: Caste pay gaps in India . The World Bank, Human Development Unit.

19.

Deshpande

(2001). Caste at birth? Redefining disparity in India. Review of Development Economics , 5(1), 130–144.

20.

Deshpande

L. K.

(1997). Gender-based discrimination in the urban labour market in India. Indian Journal of Labour Economics , 40(3), 545–562.

21.

Dickens

W. T.

(1986). Wages, employment and the threat of collective action by workers (NBER Working Paper No. 1856). National Bureau of Economic Research.

22.

Divakaran

(1996). Gender based wage and job discrimination in urban India. Indian Journal of Labour Economics , 39(2), 235–257.

23.

Duraisamy

(2005). Regional differences in wage premia and returns to education by gender in India. The Indian Journal of Labour Economics , 48(2), 335–347.

24.

Edgeworth

F. Y.

(1922). Equal pay to men and women for equal work. Economic Journal , 32(128), 431–457.

25.

Gaiha

Thapa

Imai

Kulkarni

V. S.

(2007). Disparity, deprivation and discrimination in rural India (BWPI Working Paper 13). Brooks World Poverty Institute, University of Manchester.

26.

Gandhi

Sahoo

B. K.

Saha

Mehrotra

(2013).Turnaround in India’s employment story: Silver lining amongst joblessness and informalisation? Economic and Political Weekly , 48(35), 87–96.

27.

Government of India (GoI). (2007). Report on condition of workers and promotion of livelihood in the unorganized sector . National Commission for Enterprises in the Unorganized Sector.

28.

Government of India (GoI). (2012). Report of the committee on unorganized sector statistics . National Statistical Commission.

29.

Greenhalgh

(1980). Male–female differentials in Great Britain: Is marriage an equal opportunity. Economic Journal , 90(360), 751–775.

30.

Gunderson

(1989). Male–female wage differentials and policy responses. Journal of Economic Literature , 27(1), 46–72.

31.

Ito

(2009). Caste discrimination and transaction costs in the labor market: Evidence from rural north India. Journal of Development Economics , 88(2), 292–300.

32.

Katz

L. F.

(1986). Efficiency wage theories: A partial evaluation. NBER Macroeconomics Annual , 1, 235–276.

33.

Kijima

(2006). Caste and tribe inequality: Evidence from India, 1983–1999. Economic Development and Cultural Change , 54(2), 369–404.

34.

Kingdon

G. G.

(1997). Laborforce participation, returns to education, and sex discrimination in India. The Indian Journal of Labour Economics , 40(3), 507–526.

35.

Kingdon

G. G.

(1998). Does the labour market explain lower female schooling in India. Journal of Development Study , 35(1), 39–65.

36.

Kingdon

G. G.

Unni

(2001). Education and women’s labour market outcomes in India. Education Economics , 9(2), 173–195.

37.

Kumar

Mishra

(2008). Trade liberalization and wage inequality: Evidence from India. Review of Development Economics , 12(2), 291–311.

38.

Madheswaran

Attewell

(2007). Caste discrimination in the Indian urban labor market: Evidence from the National Sample Survey. Economic and Political Weekly , 42(41), 4146–4153.

39.

Mincer

(1958). Investment in human capital and personal income distribution. Journal of Political Economy , 66(4), 281–302.

40.

Mincer

(1974). Schooling, experience and earnings (1st ed.). Columbia University Press.

41.

Mitra

(2016). Why wage differences exist across sectors? Explanations based on India’s industrial sector. Economic and Political Weekly , 54(38), 39–45.

42.

National Commission for Enterprises in the Unorganized Sector. (2007). Report on the condition of work and promotion of livelihoods in the unorganized sector. http://dcmsme.gov.in/Condition_of_workers_sep_2007.pdf

43.

Neumark

(1988). Employers’ discriminatory behavior and the estimation of wage discrimination. Journal of Human Resources , 23(4), 279–295.

44.

Oaxaca

R. L.

(1973). Male–female wage differentials in the urban labour market. International Economic Review , 14, 693–709.

45.

Pack

Todaro

(1969). Technological transfer, labour absorption, and economic development. Oxford Economic Papers , 21(3), 395–403.

46.

Papola

T. S.

(2013). Role of labour regulation and reforms in India: Country case study on labour market segmentation (Employment Sector Working Paper 147) International Labour Organisation (ILO).

47.

Phelps

E. S.

(1972). The statistical theory of racism and sexism. American Economic Review , 62(4), 659–661.

48.

Reimer

C. W.

(1983). Labour market discrimination against Hispanic and black men. Review of Economics and Statistics , 65(4), 570–579.

49.

Sengupta

Das

(2014). Gender wage discrimination across social and religious groups in India estimates with unit level data. Economic and Political Weekly , 49(21), 71–76.

50.

Singhari

Madheswaran

(2017). Wage structure and wage differentials in formaland informal sectors in Indiaevidence from NSS data. Indian Journal of Labour Economics , 60(3), 389–414.

51.

Srija

Shirke

S. V.

(2014). An analysis of the informal labour market in India (Confederation of Indian Industry, CII Working Paper). https://ies.gov.in/pdfs/CII%20EM-october-2014.pdf

52.

Srivastava

(2016). Structural change and non-standard forms of employment in India (Conditions of Work and Employment Series, ILO Working Paper 68). ILO.

53.

Stiglitz

J. E.

(1986). The wage-productivity hypothesis: Its economic consequences and policy implications (NBER Working Papers 1976). NBER.

54.

Wadhawa

K. K.

(1975). Minority safeguards in India . Thomson Press.

	Employment Type
	Informal	Formal
Sector	2004–2005
Unorganized	99.6	0.40
Organized	46.6	53.4
Total	92.4	7.60
	2011–2012
Unorganized	99.5	0.50
Organized	58.3	43.4
Total	92.9	6.90

	Formal Sector—NCEUS Definition		Formal Sector—Alternative (official) Definition
	Regular Workers		Regular Workers
	2004–2005	2011–2012	2004–2005	2011–2012
Primary	26.9	27.6	23.4	22.2
Secondary	60.2	69.7	58.2	68.9
Tertiary	31.7	41.2	26.6	35.6
Non-agriculture	40.3	50.3	36.1	46.4
Total	40.3	50.2	36.1	46.7

	Description of the Variables	Formal Workers		Informal Workers
Variables		Mean	SD	Mean	SD
lndwage	Natural Log of daily wage (₹)	6.06	0.89	5.22	0.81
age	Age in years	38.10	11.13	33.96	11.74
Age	Age square (in years)	1575.42	883.80	1291.10	893.02
Pexp	Potential experience of the worker	21.71	12.45	20.75	12.80
dBP	If the worker has below primary level of education = 1; 0 otherwise	0.16	0.37	0.32	0.47
dPrim	If the worker has primary level of education = 1; 0 otherwise	0.09	0.28	0.13	0.34
dMid	If the worker has middle level of education = 1; 0 otherwise	0.09	0.29	0.17	0.37
dSec	If the worker has secondary level of education = 1; 0 otherwise	0.14	0.34	0.13	0.34
dhsec	If the worker has higher secondary level of education = 1; 0 otherwise	0.12	0.32	0.09	0.29
ddedu	If the worker has diploma degree = 1; 0 otherwise	0.04	0.19	0.03	0.18
dgaa	If the worker has a graduate and above degree = 1; 0 otherwise	0.24	0.43	0.09	0.29
Technical education	If the worker has a technical degree = 1; 0 otherwise	0.14	0.34	0.06	0.24
Voc. training	If the worker has any vocational training = 1; 0 otherwise	0.35	0.48	0.39	0.49
Male	If the worker’s sex is male = 1; 0 otherwise	0.56	0.50	0.61	0.49
Urban	If the worker’s place of residence is urban = 1; 0 otherwise	0.69	0.46	0.67	0.47
Caste	If the worker belongs to SC or ST = 1; 0 otherwise	0.81	0.39	0.79	0.41
Married	If the worker is married = 1; 0 otherwise	0.55	0.50	0.50	0.50
Muslim	If the worker follows the Islam religion = 1; 0 otherwise	0.05	0.21	0.12	0.33
Other	If the worker follows any other religion = 1; 0 otherwise	0.07	0.26	0.06	0.24
North	If the worker is working in the north zone = 1; 0 otherwise	0.15	0.36	0.14	0.35
Central	If the worker is working in the central zone = 1; 0 otherwise	0.14	0.35	0.10	0.30
East	If the worker is working in the east zone = 1; 0 otherwise	0.16	0.37	0.11	0.31
West	If the worker is working in the west zone = 1; 0 otherwise	0.28	0.45	0.26	0.44
South	If the worker is working in south zone = 1; 0 otherwise	0.25	0.43	0.37	0.48
Agriculture	If the worker is working in agriculture = 1; 0 otherwise	0.02	0.15	0.02	0.14
Manufacturing	If the worker is working in manufacturing = 1; 0 otherwise	0.18	0.38	0.32	0.47
Services	If the worker is working in services = 1; 0 otherwise	0.80	0.40	0.66	0.47