Is growth improving employment quality in India? Evidence of widening subnational inequality

Abstract

One of the biggest challenges faced by India today is to generate quality employment. Even in areas of rapid per capita income growth, typified by the Gujarat model of neoliberal state-sponsored technological development, there is a substantial and increasing decent work deficit. Across India more generally, the decent work deficit is, in fact, growing along several dimensions, leading to ‘growth without development’ or ‘non-inclusive growth’. This article analyses quality of employment in India across subnational spaces – among states and between rural and urban locations – using three International Labour Organisation decent work dimensions: ‘employment opportunity’, ‘social security benefits’ and ‘social dialogue’. The analysis is based on published government data, for the period 1993–1994 to 2011–2012 – the period covered by the liberalisation experiment. The conclusion is that economic growth has not contributed significantly to employment quality. Although employment opportunity is significantly higher in the developed states, coverage of social security benefits and scope for social dialogue among regular salaried/wage workers are significantly less in these areas than in underdeveloped regions. Indeed, employment opportunity is significantly higher in rural areas, and the condition of workers in urban areas is not significantly better than in rural locations. Furthermore, over time, the difference in quality of employment across subnational spaces has either increased or remained stagnant.

Introduction

One of the biggest challenges faced by India today is to generate quality employment. Although economic growth has accelerated in the post-liberalisation period, and there has been an improvement in socio-economic indicators like poverty and literacy (Ministry of Statistics and Programme Implementation, 2013), their contribution towards employment, particularly quality employment, has been poor. The problem has been highlighted in recent debate over the ‘Gujarat model’ of development, adopted between 2002–2003 and 2011–2012 during Narendra Modi’s term as the state’s chief minister, and vigorously promoted by neoclassical economists (Bhagwati and Panagariya, 2012). A ‘quantum jump’ in Gujarat’s growth rate (Baskar, 2018) was achieved through policies designed to make the state the fastest growing regional economy in Asia. Based on government infrastructure spending and subsidies to corporate and foreign investors in capital-intensive technologies, the economy grew, but it is now widely recognised that this support occurred at the expense of medium and small industry, and of social investment in education and health (Baskar, 2018; Hirway, 2014, 2017a; Kannan, 2015, 2017). Hirway (2017b) argues that what emerged was a ‘highly dualistic economy with a small high-tech, high-income sector and a large informal and traditional sector’, and that the state’s export orientation made it particularly vulnerable to the global economic slowdown. After 2012, drought, environmental degradation and the aftermath of the 2010 dismantling of some agricultural protections generated large-scale migration to urban areas, swelling the informal urban workforce (Hirway, 2017b; Jaffrelot, 2015). Both Jaffrelot (2015) and Kannan (2015, 2017) describe the Gujarat model in terms of ‘non-inclusive growth’, characterised by increasing economic and social inequality. Hirway (2014) and Jaffrelot (2015) document an erosion of labour rights, and the creation of an increasing informal workforce consisting of insecure internal migrant and local labour, in which women and minorities are particularly disadvantaged. The question addressed in this article is whether this model of increasing inequality applies more widely in India.

Kannan (2015) argues that the main basis of poverty reduction is through the quality of employment, wages and conditions of work, including the availability of social security. This article uses decent work indicators to explore the extent to which the increasing inequality accompanying growth in Gujarat during the period of neoliberalisation was a microcosm of unequal development across states and subregions of India generally between 1993 and 2012. It finds a substantial decent work deficit (Institute for Human Development (IHD), 2014; National Commission for Enterprises in the Unorganised Sector (NCEUS), 2009) and shows that this deficit in fact grew along several dimensions during the two decades studied. If employment opportunity is measured by workforce participation rate and unemployment rate, the overall situation has deteriorated. The majority of the Indian workforce is engaged in the informal sector and only a small segment is employed in formal activities; moreover, the predominance of the informal sector actually grew during the period studied. Within the informal sector, there is a very large section involved in self-employment-based activities. Not only most of the informal sector but also a large segment of the formal sector remains dependent on casual work, and the casual labour force has almost no work-related forms of security. Only a very small fraction of the Indian workforce enjoys regular work status with access to work-related social security. Both for the informal sector in general, and for the pool of self-employed and casual workers in particular, the conditions of employment were very poor during the period in question, and have not substantially improved since then, despite overall economic growth. If we look more deeply at crucial dimensions of quality of employment – coverage of social security benefits, scope for social dialogue, stability and security of jobs, and equal opportunity and treatment in employment – the condition is not only miserable but, in some regions and aspects, has even been deteriorating.

Using detailed regional data analysis for the post-liberalisation period up to 2012, this article establishes a divergence of quality of employment across subnational states, and for rural and urban locations. Even in those states which were performing better in terms of economic growth, it will be argued that there was not a commensurate improvement in labour conditions. Instances will be identified where economic growth bore an inverse relationship to overall quality of employment. Curiously, the more economically developed states will be shown to have performed poorly on certain equality measures, and some interstate inequalities in employment quality will be shown to have increased. Yet to date, policy prescriptions have focused on the quantitative aspects of job generation, ignoring qualitative considerations (Moktan, 2016).

The findings were derived by the following approach: (a) comparative use of certain dimensions and indicators of employment quality, (b) identification of variations in employment quality by states as a way of indication differences between rural and urban locations and (c) tracing the changing patterns for aspects of employment quality over time. Three dimensions of employment quality were used, based on International Labour Organisation (ILO) criteria: employment opportunity, coverage of social security benefits and scope for social dialogue (ILO, 2012). These dimensions were chosen because of the availability of tangible variables for India. Data on these variables were collected for three post-liberalisation periods during the years 1993–1994 to 2011–2012.

Despite lack of a panel data set or use of household-level information, this article’s interstate analysis across rural–urban locations for the post-liberalisation period should add value to the existing literature. There are not many studies which consider the variety of dimensions of Quality of Employment for India at comparative state and regional level, and for several post-liberalisation phases. This study may therefore be viewed as an initial step in understanding the variations in quality of employment across subnational Indian states, their rural–urban locations and, over time, with the help of dimensions as proposed by the ILO.

For the three dimensions, I selected key indicators first and then chose variables according to the availability of data in India. As explained more fully in the ‘Methodology’ section, the states were divided into three categories according to their economic conditions: developed, moderately developed and underdeveloped states. In order to measure the extent of variation in quality of employment, the all-India level (including all states and union territories) was kept as a yardstick. Mean difference tests were used to identify variations across states and rural/urban locations.

The main findings are as follows: (a) quality of employment in terms of employment opportunity was higher in the economically developed states than in underdeveloped states; (b) nevertheless, in terms of dimensions such as coverage of social security benefits and scope for social dialogue, the developed states were performing poorly in comparison to other states; and (c) not only did a rural–urban gap exist but also these gaps widened over time for most indicators. As research in this area is scanty, this study may be viewed as an initial step in understanding the variations in quality of employment across subnational Indian states, their rural–urban locations and, over time, with the help of dimensions as proposed by ILO.

The article is organised in five sections. In the ‘Literature review’ section, a literature review is used to identify knowledge gaps. In the ‘Methodology’ section, the research methodology is discussed in detail. In the ‘Findings (general trends and empirical tests)’ and the ‘Discussion’ sections, findings for data analyses are reported and discussions of the study are given, where we find the diverging quality of employment across subnational spaces. Finally, the ‘Conclusion’ section discusses the implications and significance of the study, presenting a critique of the current concept of ‘inclusive growth’, by showing that quality of employment has not improved with economic growth.

Literature review

The ILO (2014) has observed that decent job creation has slowed down almost everywhere in the world. In developing countries, most workers are working informally in vulnerable conditions, without access to social protection or secure job contracts and with poor wages (ILO, 2014). In addition, despite growth in developing countries, the capacity to generate stable and remunerative employment, even in formal activities, is limited (Ocampo, 2008; United Nations Development Programme (UNDP), 2013).

In India, there is a deficit in quality of employment (NCEUS, 2009). The majority of workers are still to be found in sectors with low productivity and earning capacity, limited or no social protection and limited access to formal job contracts offering paid leave and other indirect pay benefits (IHD, 2014; NCEUS, 2009; Papola and Sahu, 2012). The IHD (2014) Annual Employment report observed that workers’ conditions in the poorer states were worse in terms of regularity, stability and security. On the other hand, in the economically developed states of Gujarat and Andhra Pradesh, workers were engaged largely as informal workers, lacking labour market or work security, and with low labour productivity and wages (Dev and Mahajan, 2003; Kantor et al., 2006). By December 2017, this was still the case: Hirway (2017b) argued that over the preceding decade, Gujarat had been the only state where the formal sector actually declined as a proportion of the overall workforce – despite a workforce participation rate second only to Haryana’s, employment quality was consistently poor, with 94% in the informal sector, subject to low wages and poor access to social security. She cited evidence that even in the formal sector, unskilled workers in Gujarat were earning only a little over three-quarters of the national average, and that overall there was a severe shortage of decent jobs for young people seeking to enter the labour market. Across India as a whole, Verick and Chaudhary (2016: 1) noted that the ‘vast majority’ of workers were still in informal jobs, with construction work absorbing many of those moving out of agriculture: ‘most of the new jobs being created in the formal sector are actually informal because the workers do not have access to employment benefits or social security’ (Verick and Chaudhary, 2016: 1).

To measure quality of employment, developing countries rely on indicators such as wages, job safety, security benefits, unions and working hours (Dewan and Peek, 2007; Flanagan and Khor, 2012; United Nations Conference on Trade and Development (UNCTAD), 2013). In India, most studies on employment quality are based on a limited number of dimensions, particularly average wages and regularisation of employment contracts (Kantor et al., 2006; Papola and Sahu, 2012; Unni and Raveendran, 2007). So, a study on quality of employment including other dimensions is a worthwhile addition to the literature. Studies on quality of employment across subnational spaces such as across states and rural–urban locations are limited. Besides, over time, analyses of variations across subnational spaces are also sparse in India.

In this context, the present study is based on ILO dimensions and indicators of decent work. The idea of ‘decent work’ was first articulated in 1999 by the ILO Director-General in his report to the 87th Session of the International Labour Conference (ILO, 2008: 4). He described decent work in terms of ‘opportunities for women and men to obtain decent and productive work in conditions of freedom, equity, security and human dignity’ (ILO, 2008: 4). This framework covers 10 substantive elements: employment opportunities; adequate earnings and productive work; decent working time; combining work, family and personal life; work that should be abolished; stability and security of work; equal opportunity and treatment in employment; safe work environment; social security; and social dialogue, and employers’ and workers’ representation (ILO, 2012: 7). Out of this list, I have focused on three dimensions: employment opportunities, social security and social dialogue.

Employment opportunity targets an important aspect of labour market conditions faced by workers and employers. This dimension ‘includes indicators which permit the analysis of quality of employment measured through the lens of informal employment and other key components of total employment’ (ILO, 2012: 45).

‘Social security’ covers:

all the measures that provide benefits, whether in cash or in kind, to secure protection, inter alia, from lack of work-related income (or insufficient income) caused by sickness, disability, maternity, employment injury, unemployment, old age, or death of a family member; lack of access or unaffordable access to health care; insufficient family support, particularly for children and adult dependants; general poverty and social exclusion. (ILO, 2012: 148)

‘Social dialogue’ covers:

all types of negotiation, consultation, and exchange of information between representatives of governments, employers and workers on issues of common interest. It covers both tripartite processes and institutions of social dialogue, such as social and economic councils; institutions, such as trade unions and employers’ organisations; and processes, such as collective bargaining. (ILO, 2012: 164)

The dimensions, indicators and variables chosen for the study are set out in Table 1. In order to measure these dimensions, certain indicators are provided by the ILO. We have chosen some of these important indicators based on the study’s objectives: these indicators are set out as categories in column 3 of Table 1. Finally, on the basis of these categories, I have drawn out the variables to be studied according to the availability of data.

Table 1. Dimensions, indicators, categories and variables used.

Source: ILO (2012: Table No-A, pp. 16–17).

ILO: International Labour Organisation; PF: provident fund; AGEGC: [ag]ricultural sector [e]xcluding only [g]rowing of [c]rops, market gardening, horticulture and growing of crops combined with farming of animals.

Methodology

This study is based on published data from the Government of India’s National Sample Survey Office (NSSO). In what are called quinquennial rounds, large-scale sample surveys are conducted over almost the entire country every 5 years. The sample size in each year consists of data from around 400,000 households. These NSSO data have been used mainly for four quinquennial rounds: 50th (1993–1994), 55th (1999–2000), 61st (2004–2005) and 68th (2011–2012). The analysis also uses the Government of India’s 2001 and 2011 population censuses. Thus, detailed data are available for rigorous analysis to 2012; in the conclusion, I refer to less comprehensive evidence of the ongoing relevance of the findings.

As explained above, for the purposes of this study, the full range of ILO dimensions is narrowed down to three, as follows.

First, ‘employment opportunity’ is measured in terms of (a) workforce participation rate, (b) unemployment rate, (c) the ratio of total number of informal workers to total workers ratio and (d) different categories of employment (self-employed, regular and casual workers). The first two indicators give us an idea about labour market conditions, while the other two indicate in a nutshell the nature of employment.

Second, in order to measure the coverage of social security benefits, we use the proportion of those who are deprived of social security benefits¹ as a percentage share of total workers engaged in the AGEGC ([ag]ricultural sector [e]xcluding only [g]rowing of [c]rops, market gardening, horticulture and growing of crops combined with farming of animals) and non-agricultural sectors.

Third, the scope for social dialogue is measured as a percentage share of workers who are working without any unions/associations² in their activity.

We use usual activity status³ (principal + subsidiary status) data for all. In measuring decent work in terms of coverage of social security benefits, the analysis is confined to workers engaged in the AGEGC and non-agricultural sectors, owing to non-availability of data on all the workers. However, 41% of workers in rural areas and 95% of workers in urban areas were engaged in the AGEGC and non-agricultural sectors (NSSO, 2014b: ii).

All the three dimensions were studied across 15 major states⁴ and their rural–urban locations. For the analysis, I categorised states with Net State Domestic Product (NSDP)⁵ per capita (pc) (at constant 1993–1994 prices) in 2005–2006 as developed (NSDP pc > Rs. 16,000), states with NSDP pc < Rs. 12,000 as underdeveloped and states in between as moderately developed (Rs. 12,000 ⩽ NSDP pc ⩽ Rs. 16,000). According to these criteria, states like Gujarat, Haryana, Maharashtra, Punjab and Tamil Nadu were classified as developed states; Assam, Bihar, Madhya Pradesh, Orissa, Rajasthan and Uttar Pradesh as underdeveloped states; and Andhra Pradesh, Karnataka, Kerala and West Bengal as moderately developed states. Data were analysed by taking the mean value of the states’ NSDP pc in their particular groups. The all-India level (including all states and union territories) was kept as a yardstick to compare decent work outcomes. Mean difference tests were also carried out for some indicators to see whether there are any differences across states and their rural–urban locations.

Findings (general trends and empirical tests)

Employment opportunity

Workforce participation rate

In India, in both rural and urban locations, on average, less than half of the total population was found to be working, and participation rates declined over the time frame studied (Figure 1 and Supplemental Tables S1-A and S1-B). Although the participation rate was higher in rural areas than urban areas, between 1993–1994 and 2011–2012, with the exception of 2004–2005, participation rates declined in rural areas but remained almost constant (with some fluctuations) in urban areas.

Figure 1. Rural and urban workforce participation rates by states’ level of development, India, 1993–1994 and 2011–2012.

Source: National Sample Survey Office (NSSO, 1997, 2001a, 2006, 2014a).

Comparing different states, Figure 1 and Supplemental Tables S1-A and S1-B indicate that participation rates in rural areas were higher than the all-India level in developed and moderately developed states but lower in underdeveloped areas. This was true even in the urban areas, except in 1993–1994. However, over time, except in 2004–2005, participation rates declined in all groups of states in rural areas. In urban areas, with some fluctuations, participation increased in developed states, remained almost constant in moderately developed states and slightly declined in underdeveloped states. The results of a mean difference test involving workforce participation rates across rural and urban locations (Appendix 1, Table 2) suggest that employment opportunity in terms of workforce participation rate was significantly higher in rural areas than urban areas. At the same time, mean difference tests between developed and underdeveloped states, and between moderately developed and underdeveloped states (Appendix 1, Tables 3 and 4), indicate that employment opportunity, measured by workforce participation rate, was significantly higher in the economically developed and moderately developed states than in underdeveloped areas.

Unemployment rate

The all-India unemployment rate was higher in urban areas than rural areas (Figure 2 and Supplemental Table S2). However, over time, the rate slightly increased in rural areas. In urban areas, although the rate declined, it remained higher than in rural areas.

Figure 2. Rural and urban unemployment rates by state’s level of development, India, 1993–1994 and 2011–2012.

Source: National Sample Survey Office (NSSO, 1997, 2001a, 2006, 2014a).

Comparing different states, we see that unemployment rates in both locations (rural and urban) are higher than the all-India level in moderate and underdeveloped states. Over time, the rate showed an increasing trend till 2004–2005, and then declined in both locations in all groups of states irrespective of economic status. However, compared to 1993–1994, in 2011–2012, the rate increased in rural areas in all groups of states but decreased in urban areas.

Informal employment

The majority of workers in the non-agricultural sector excluding construction⁶ were engaged as informal workers (Figure 3 and Supplemental Table S3). Moreover, the majority of the agricultural sector and construction were also comprised of informal workers. In both the formal and informal sectors, informalisation (measured as the ratio of informal workers to total workers) increased over time. While the number of informal workers in the non-agricultural sector was higher in urban than rural areas, over time this ratio increased in rural areas, while it slightly declined in urban locations.

Figure 3. Ratio of informal workers to total workers (excluding agriculture and construction) – rural and urban by level of development, India, 1999–2000 and 2010–2011.

Source: National Sample Survey Office (NSSO, 2001b, 2012); Office of the Registrar General & Census Commissioner, India (2011).

This ratio of informal workers to total workers was higher than the all-India level for the rural areas of moderately developed states and for the urban areas of developed states. Most disturbingly, it increased in the rural areas of developed and moderately developed states. For urban areas, it remained constant in developed states while declining in other locations.

Employment status

In terms of employment category (self-employed, casual and regular salaried/wage workers), in rural areas, more than half of the workers were engaged as self-employed, about one-third as casual workers and the rest as regular workers (Figures 4 and 5; Supplemental Tables S4 and S5). In urban areas, about two-fifths of workers were self-employed, two-fifths were regular workers and the rest were casual workers.

Figure 4. Percentages of workers in formal and informal employment categories – rural areas by state’s level of development, India, 1993–1994 and 2011–2012.

Source: National Sample Survey Office (NSSO, 1997, 2001a, 2006, 2014a).

Figure 5. Percentages of workers in formal and informal employment categories – urban areas by state’s level of development, India, 1993–1994 and 2011–2012.

Source: National Sample Survey Office (NSSO, 1997, 2001a, 2006, 2014a).

Over time, in rural areas, the share of self-employed declined while that of casual workers increased (except in 2004–2005), whereas the share of regular workers increased continuously. In urban areas (with some increase in 2004–2005), self-employed workers’ share remained almost constant, casual workers’ share remained almost constant (with some decline in 2004–2005) and regular workers’ share remained almost constant till 2004–2005, and increased in recent years.

Comparing states using Figures 4 and 5 (and Supplemental Tables S4 and S5), we see that the share of self-employed was higher than the all-India level for both rural and urban areas of underdeveloped states. The shares of regular and casual workers were higher in rural areas of both developed and moderate states. In urban areas, the shares of regular workers were higher in developed states, and of casual workers in moderately developed states. Over time, for rural areas, the share of regular workers increased, particularly at a higher rate in developed and moderately developed states. The share of self-employed declined in both of these states (except in 2004–2005), but it remained constant overall in underdeveloped states, with some minor fluctuation. The share of casual workers declined in developed and underdeveloped states but increased slightly in moderately developed states, with some fluctuations. For urban areas, with some fluctuation, the share of self-employed declined, the share of regular workers has increased and that of the casual workers declined continuously in the developed and moderate states. However, in underdeveloped states, again with some fluctuations, the share of self-employed increased, the share of regular workers decreased and that of casual workers remained almost constant.

Social security benefits

Almost all casual workers and more than half of regular workers lacked social security benefits, in both rural and urban locations (Figure 6 and Supplemental Table S6). While there were no rural–urban differences in deprivation for casual workers, the shares of regular workers deprived of these benefits were higher in rural than urban locations. Furthermore, these rural–urban gaps widened as the share of workers deprived of these benefits increased in rural areas, while remaining almost constant in urban locations.

Figure 6. Percentages of workers lacking social security benefits, by state’s level of development, India, 2004–2005 and 2011–2012.

Source: National Sample Survey Office (NSSO, 2007, 2014b).

When comparing states, we see that the share of casual workers deprived of social security benefits was higher than the all-India level for developed and moderately developed states in both locations (rural and urban). For regular workers, the share was higher in developed and moderate states in rural areas, and higher only in the developed states in urban areas. Furthermore, the share of regular workers increased in rural areas of moderately developed and underdeveloped states, and remained constant in developed states. For urban areas, the share declined in moderately developed states but increased in developed and underdeveloped states. However, interestingly, the share of casual workers for both areas decreased among all three groups of states, irrespective of economic status.

The mean difference tests across the rural and urban locations (Appendix 1, Table 2) for regular and casual workers suggest that there was no rural–urban gap in the coverage of social security benefits. However, the tests between the developed and underdeveloped and between moderate and underdeveloped states (Appendix 1, Tables 3 and 4) show significant differences. This result suggests that coverage of social security benefits among both regular and casual workers was significantly lower in the economically developed states than the underdeveloped areas.

Social dialogue

When we turn to scope for social dialogue, we see that the majority of self-employed and casual workers, and more than half of all regular workers, had no representation by, or participation in, a union or other collective organisation, a concept represented by the variable, ‘non-existence of a union/association in their activity’.⁷ Figure 7 and Supplemental Table S7 indicate that the share of workers without a union in their activity was higher in rural than in urban locations for all categories of workers. Disturbingly, over time, the proportion lacking collective engagement has increased among all categories of workers in both rural and urban locations.

Figure 7. Percentages of workers without any union/association in their activity by employment status and state’s level of development, India, 2004–2005 and 2011–2012.

Source: National Sample Survey Office (NSSO, 2007, 2014b).

In comparing states, it can be seen that the share of regular workers without any union/association in their activity was higher than the all-India level for both rural and urban locations in the developed states. The shares of casual workers without union involvement were higher for both rural and urban areas of both developed and underdeveloped states. Similarly, concentrations of non-unionisation were highest among self-employed workers in rural areas of underdeveloped states, and urban areas of both developed and underdeveloped states. Over time, these shares increased in all groups of states, irrespective of their economic status.

The mean difference test (Appendix 1, Table 2) shows no significant difference between rural and urban areas in shares of regular and casual workers lacking union/association cover. However, for the self-employed, lack of voice was significantly higher in rural areas than urban locations. The mean difference test between developed and underdeveloped states (Appendix 1, Table 3) indicates that the voice of regular workers was significantly stronger in underdeveloped than in developed states. On the other hand, the test between moderately developed and underdeveloped states (Appendix 1, Table 4) indicates that the voices of both self-employed and casual workers were significantly stronger in moderately developed than in underdeveloped areas.

Discussion

While the study relies on a comparison of available data collected between the 50th and 68th NSSO quinquennial rounds, there is no evidence of more recent divergence from the policy directions and outcomes of these foundational years of neoliberalism. Overall, during the two decades to 2012, employment opportunity, as seen through the lens of workforce participation rates and unemployment rates, was higher in rural areas than urban areas. However, over time, opportunity deteriorated in rural areas while not improving in urban areas. For other indicators like forms of employment (e.g. presence of casual workers), coverage of social security benefits and scope for social dialogue, the condition of workers in rural areas was not only poorer than in urban areas but also deteriorated over time. The mean difference test results suggest that, although employment opportunity was significantly higher in rural than urban areas, differences for other indicators such as social security benefits and social dialogue among regular and casual workers were not significant.

At a subnational level, employment opportunity was higher in developed states than less developed ones. But the condition of workers in terms of form of employment (presence of informal and casual workers), coverage of social security benefits and scope for social dialogue was poorer in developed states than underdeveloped locations. In addition, over time, most of these indicators had either deteriorated or remained constant, particularly in developed states. Moreover, the mean difference test results also suggest that, although employment opportunity was significantly higher in developed states, the coverage of social security benefits and scope for social dialogue, particularly among regular workers, were significantly less in developed states compared to underdeveloped areas.

Over the period studied, employment opportunity, indicated by workforce participation and unemployment rates, was higher in rural than urban areas. This was mainly because people in rural areas were largely engaged in agriculture, where entry barriers are low. Employment opportunity was higher in developed and moderately developed states. This can also be inferred from higher NSDP per capita from agriculture in the developed and moderate states in comparison to the underdeveloped ones. However, over time, employment opportunities in rural areas declined, while remaining constant in urban areas. Since the inception of new economic reforms, agriculture has been neglected, with the focus being on the manufacturing and service sectors. However, there are entry barriers in these sectors in terms of education, gender and age. There has also been a decline in operational landholding in almost all states, particularly owing to population pressure, which has led to a decrease in demand for workers in the farm sector. Third, there has been an increase in capital-intensive methods of production over time in both the farm and non-farm sectors (Papola and Sahu, 2012).

Comparing states, we see that employment opportunity declined over time in all groups of states in rural areas. However, while employment increased for urban areas in developed states, it remained constant in moderately developed states. The increase in employment opportunity in developed states could be due to increases in infrastructural activities, such as construction: the share of NSDP from services and construction was higher in developed states than others (Ministry of Finance, Government of India, 2018).

The rise in the workforce participation rate in 2004–2005 (in comparison to the previous period) has to be seen in context. The 1999–2005 period was marked by a serious agrarian crisis in India. The decline in agricultural growth or yield had adverse effects on food supply, prices of food grains, cost of living, poverty, employment and so on. Scholars argue further that, if there is a perceived fall in household income, individuals – particularly females – tend to enter the labour force. In times of distress, there is a rise in both the labour force participation rate and the unemployment rate (Abraham, 2009, 2013; Himanshu, 2011). In fact, there was a rise in these variables during this period. Scholars have termed it the ‘income effect’. So, although participation rates increased, the question arises regarding the condition of workers.

Turning to the nature of employment, we see not only that the majority of workers were engaged in the informal sector but also that the proportion of informal workers (in both the formal and informal sectors) increased over time (Heintz and Pollin, 2003; ILO, 2002; NCEUS, 2009; Papola and Sahu, 2012). Moreover, a large number of workers in the informal sector were engaged in the self-employment-based own account segment,⁸ which has lower productivity (per enterprise and per worker) compared with the formal establishment⁹ sector (NSSO, 2001b: iii, 2012: 25). Such workers lacked entitlement to the government-stipulated minimum wage. This implies poor quality of employment. In addition, informal workers are largely engaged in irregular and insecure jobs, without work-related security or proper workplaces, and with low productivity and earnings (ILO, 2002; NCEUS, 2009; Papola and Sahu, 2012). All these factors imply that the majority of workers were engaged in poor quality employment.

The higher number of informal workers in economically developed states than in underdeveloped ones implies that, despite economic progress, workers’ condition remained poor. The nature of employment opportunity, in terms of forms of employment, further indicates that the majority of workers had limited avenues to work, as self-employed and casual workers, particularly in rural areas. Regular workers had more privileges than casual in terms of coverage of social security, scope for social dialogue and stability of work. However, the self-employed are a mixed category, including those working on their own large farms and in enterprises with high income, as well as own account workers on tiny farms and in enterprises with low income.

Over time, in rural areas, with some fluctuations, while the share of regular workers increased continuously, the share of self-employed decreased and that of casual workers increased. The reason may be a secular decline in operational landholding and an increase in the number of landless farmers, leading to a fall in self-employment and a rise in the casual workforce. In addition, the non-farm sector has been growing in rural areas, which could also have led to the rise in casual as well as regular workers.

A further reason for casual workforce growth, particularly after 2005, could be public works programmes undertaken by the government of India, in particular those introduced under the Mahatma Gandhi National Rural Employment Guarantee Act (MGNREGA).¹⁰ Reportedly, on average, 5 crore (50 million) households were provided employment every year from 2008 (Ministry of Rural Development, Government of India, 2012). As far as the urban areas are concerned, we see that the condition of workers in terms of casualisation was (and remains) poor in underdeveloped states, as expected.

Lack of coverage of social security benefits and limited scope for social dialogue impact the quality of work life particularly of self-employed and casual workers, but over half of regular workers, too, lack social security benefits, and work in a non-union environment. This is mainly due to the informal sector’s dominance, making it difficult to enforce rules and regulations (such as the Workmen Compensation Act, the Maternity Benefit Act, the Building and Other Construction Workers Act; ILO, 2002; NCEUS, 2009). In addition, both the central and the state governments have made only limited efforts to introduce social security schemes for the informal sector (IHD, 2014: 148).

Furthermore, a large number of workers in the informal sector are engaged in those sub-sectors which have lower productivity. The limited earnings of informal sector employers also could limit social security benefits to the workers. Besides these, a huge number of workers lack voice at their workplaces, contributing to poorer social security benefits. This is also the case for developed and moderate states. Dominance of the informal sector helps explain the lack of social dialogue. The number of small enterprises, dominance of self-employment and casual employment, and widespread worker illiteracy all pose major challenges in forming unions (IHD, 2014). Large numbers of workers lack stability and security of work – they are engaged as temporary workers and are working without job contracts (NSSO, 2007: 107–109, 2014b: 138–141). Combined with the lack of employment opportunity, these factors inhibit workers from forming and joining unions, through fear of losing jobs. The share of non-unionised workers is larger in rural areas than urban, presumably because the majority of rural workers are still engaged in agricultural activities, where it is difficult to form unions. In addition, the number of small enterprises, dominance of self-employment and casual employment, illiteracy rate, lack of stability and security of work are all much higher in rural areas than urban locations.

The dearth of unions in developed states relative to others could also be due to the larger number of informal workers in these states than others. However, although enterprise size, literacy rate and employment opportunity are higher in developed states than underdeveloped areas, low unionisation rates, particularly in urban areas, could be due to the increase in new sectors like information technology (IT), IT-enabled services (ITES), hospitality and new retail establishments like shopping malls, which face problems in unionising (IHD, 2014).

Overall, despite economic growth, the condition of labour has not improved with economic progress. While employment has increased in the economically developed states, these states have performed worse than poorer states in terms of coverage of social security benefits and scope of social dialogue. An explanation is that growth in employment, even in economically developed states, has been of the informal kind, even within the formal sector. Moreover, large sections of workers in the informal sector were (and are) working in sub-sectors with low productivity. In addition, one of the major problems in the informal sector is that, though regulations are framed by the government, implementation and enforcement of these laws are limited and ineffective. This may further indicate that economic growth alone is not sufficient to improve quality of employment. State intervention is essential to ensure employment quality, as the market cannot take care of these constraints on its own. Moreover, the level of education and skills are important for improving the condition of workers. Education would increase knowledge and awareness about individual rights, while skills can increase productivity per worker, further enhancing quality of employment.

Conclusion

The main purpose of this study has been to understand employment quality in India across subnational spaces, such as states and rural–urban areas, with the help of indicators developed by ILO. This issue assumes importance in a context where, although the Indian economy has been experiencing high growth, the contribution of this growth towards employment is poor. The article has focused on three important dimensions of quality of employment proposed by the ILO – employment opportunity, social security benefits and social dialogue. While the evidence considered is for the period 1993–2012, there is no reason to believe that there has been any change in the years that followed. As Kannan (2015) has noted, the policies underpinning the Gujarat model have since been emulated in other states and have been carried across into the politics of the country as a whole.

It is safe to conclude that not only is there a lack of employment opportunity but also those workers who have been gaining jobs are engaged in the worst forms of labour, mostly in the informal sector. Within the informal sector, the majority of workers are engaged in sub-sectors with low per-worker and per-enterprise productivity. Almost all casual workers and more than half of regular workers lack social security benefits. In addition, the majority of self-employed and casual workers, and more than half of regular workers, are not associated with any union, indicating a lack of voice. Furthermore, there is a continued wide divergence among different groups of states and between rural and urban locations. Although employment opportunity is significantly higher in economically developed states in comparison to the underdeveloped ones, my analysis has indicated that both coverage of social security benefits and scope for social dialogue among regular workers were significantly less than in underdeveloped states. So far as disparities between rural and urban locations are concerned, we see that employment opportunity was significantly higher in rural areas than urban locations. For other indicators such as social security benefits and social dialogue, there was no significant difference although the condition of workers in urban areas seems to be better than in rural locations. Furthermore, over time, variations between subnational states and rural–urban locations either increased during the decades studied or did not improve, for almost all indicators.

Moreover, all these findings indicate that economic progress is perhaps not contributing in any substantial way to quality of employment. In general, economic growth and the operation of the unregulated market may not automatically lead to significantly higher quality of employment. In order to reduce the deficit in quality of employment, government intervention is necessary. In addition, a case can be made that cluster-based industrialisation (involving millions of small and micro enterprises) may also add to quality of employment through increases in economies of scale and enhanced bargaining power in both input and output markets, as well as improved productivity, knowledge and skills.

The present study may be viewed as an initial step in understanding quality of employment, and its variations across subnational Indian states and their rural–urban locations. Moreover, this study adds to our knowledge, given the limited number of works that consider a variety of dimensions of quality of employment in India. In particular, there is a paucity of studies measuring work quality at the subnational level of states and rural–urban locations. Nor have various dimensions of decent work deficits been tracked on a regional basis over several post-liberalisation phases, as has been done in the present study. This study can be considered to be a partial analysis of quality of employment, as many important dimensions of employment quality have not been addressed due to the paucity of space, as well as lack of secondary data. Nevertheless, it lays the foundation for further research, in the most recent decade, into the phenomenon of ‘growth without development’ or ‘non-inclusive growth’.