Access to Non-farm Employment in Contemporary India: A Study of Bihar and Punjab

Abstract

Rural non-farm diversification in India is taking up new roles amidst increasing agrarian distress. In this context, two issues have been examined in this paper: first, the nature of rural non-farm diversification, and second, the accessibility of households to rural non-farm employment in the states of Bihar and Punjab. The study is predominantly based on unit level data of the latest round of the Situational Assessment Survey of Agricultural Households (NSSO). Findings suggest that while non-farm activities are largely adopted by landless and marginal land households in both states, there are a few lucrative options available which are being accessed by large landholders in Punjab. Overall, caste, gender and education are dominant determinants that work as barriers to the entry for rural households. The findings recommend that institutional reforms along with public policies should be prioritised towards generating sustainable non-farm livelihood options while eliminating multi-dimensional exclusions in rural labour markets considering regional prerequisites.

Keywords

Rural non-farm employment diversification and determinants

Introduction

The deep-rooted agrarian crisis in India is the result of an accumulation of distress for more than two decades. In recent years, the crisis in agriculture has been viewed differently compared to its previous nature due to the fact that this sector is facing severe distress even when there is a continuous increase in the production of output. Despite the tremendous agricultural growth shown by economically backward states such as Rajasthan, Madhya Pradesh and Bihar there has been visible distress in rural India. However, policies to curtail agrarian crisis have remained largely confined to individual targetted schemes such as direct income benefits to farmers, loan waivers, interest subventions, crop insurance and so on do not really address the need for sector-wide improvements. Most of these measures are politically induced and provide a temporal effect to ease out the crisis (Himanshu, 2019). No focus has been laid on reviving the agrarian economy through focussed public investment on enhancing the growth of farm and non-farm economy (CBGA, 2020). The growth of non-farm employment not only eases the agrarian crisis and enhances household income but also facilitates smooth structural transformation of the rural economy (Binswanger-Mkhize, 2012; Haggblade et al., 2009; Haque, 2016). Against this backdrop, the present article attempts to explore the importance of rural non-farm employment in two-economically diverse states of India, Punjab and Bihar.

A comparative analysis of economically diverse regions will help us to understand the variety of processes that shape opportunities for households to participate in the non-farm sector. Knowing the existing livelihood options and factors affecting farmers’ participation in non-farm activities is crucial for understanding drivers of employment growth. Therefore, the intent of this article is to identify the socio-economic factors which determine the decision of smallholder farmers’ participation in non-farm activities.

This article is divided into seven sections. After the introduction, the second section provides the summary of literature on the issue of non-farm employment. While the third section explains the methodological framework and the data used for the analysis, the fourth discusses employment growth trends in rural Punjab and Bihar. The fifth section maps rural employment opportunities and the patterns across various land size groups and the sixth attempts to identify the determinants of employment diversification in the non-farm sector. The last section concludes the discussion.

Why the Revival of Rural Non-farm Sector is of Paramount Importance

Literature on the rural non-farm economy in developing countries and particularly in India has found that activities in this area are normally small-scale, require low entry capital, are seasonal in nature and are mostly home-based. However these activities plays an important role in a region’s economic transformation (Haggblade et al., 2009). Rural non-farm employment has been viewed as a viable adaptation strategy to combat agrarian crisis, climate change-induced shocks, market failure and so on. (Asfaw et al., 2017; Ellis, 1998). Households engaged in these activities are capable of withstanding an agrarian crisis and have a more stable livelihood than those which have their farms as a single source of income (Seng, 2015).

Two primary reasons for the diversification of rural households into non-farm employment have been discussed in the literature studied. Firstly, there are the push factors––when the agricultural sector is not performing adequately and is unable to provide enough employment to the farm sector. In such a scenario, workers move out of agriculture to sustain their income level and seek employment in the non-farm sector. The data from Employment and Unemployment Surveys of the National Sample Survey Office (NSSO) suggests that during the reform period of the Indian economy from 1991 onwards, the rural non-farm sector was able to absorb a large labour force. In the pre-reform period, the annual growth of farm employment was 1.24 per cent which decelerated to 0.86 per cent in the reform period whereas the growth of non-farm employment was 3.05 per cent in the pre-reform period and 3.81 per cent in the reform period (Bhaumik, 2007). These figures indicate that there has been an increasing movement of labour from the farm to the non-farm sector in the reform period.

Increasing employment growth in the non-farm sector has been explained as distress diversification by various studies (Chadha, 2003; Abraham, 2013; Vaidyanathan, 1986). Bhalla (1993) has argued that there were two types of distress diversification––the first can be seen in the case of those subsidiary workers who do not have a main occupation but are engaged in some subsidiary work; the second can be seen in the case of those who have a main occupation but are also engaged in a secondary activity. In both these cases of distress diversification, non-agricultural wages (or returns to family labour) are likely to be lower than prevailing wage rates or even below subsistence levels (Unni, 1994).

The second process is through pull factors where rural workers gain employment in the non-farm sector generated due to a better performing farm sector. In this process, there is growth in agricultural sector outflows and capital and new opportunities are generated through production and consumption linkages (Haggblade et al., 2009; Visaria & Basant, 1994). In the second process, there is an expected increase in employment and wage rates in the non-farm sector. The improvement in the rural non-farm sector is also possible through the impetus received from urban areas in terms of its spread effects (Haggblade et al., 2009; Jatav & Sen, 2013). The combined relevance of push and pull factors suggests that there are two sets of non-farm activities: firstly, those which serves as a last resort of activities for the poor, secondly, those which are profitable or more remunerative.

The movement of the workforce to the non-agricultural sector is also governed by the households’ accessibility to this sector. There are certain factors that influence accessibility such as education, wealth, size of the household, caste, village-level agricultural conditions, population densities and other regional factors (Lanjouw & Shariff, 2004). Therefore the shift of peasants from cultivation to non-farm occupations and the quality of employment depends largely upon the characteristics of households (Jha, 1997; Ellis, 1998; Harriss-White, 2014). The favourable factors ease access to non-farm opportunities while the push factors force farmers to abandon cultivation. The pull factors emerge due to attainment of higher education, availability of capital investment, a niche market for a product and wage differential; whereas the push factors are driven out by low returns in the farm sector, climate change and crop loss, higher family size, low capital-base and so on.

Although non-farm diversification is widespread in rural India not all households afford their access to lucrative non-farm employment (Asfaw et al., 2017; Barrett et al., 2001). At the regional level, the growth of rural non-farm employment depends upon the overall growth scenario of an economy. Economic conditions of regions offer different prospects of aggregate trends in farm and non-farm productivity, labour allocation across sectors, income composition and fluidity of rural-urban interactions (Haggblade et al., 2009; Jha, 2016; Tacoli, 1998).

There are ample studies on various aspects of rural non-farm employment growth in India. However, there is still a need to examine the variety of processes at the micro-level that are shaping employment opportunities for rural households at the current juncture. Secondly, there is limited available information on the sources and incomes of rural non-farm employment; rather scholars have attempted studies with primary information which are often subjected to a particular geographical and socio-economic situation. The present study uses the rich source of household-level data available in the latest round of the Situational Assessment Survey of Agricultural Households which collects information on various aspects of employment and income from a relatively large number of rural households.

Data and Methodology

Data

The study is largely based on agricultural household¹ data provided by the Situational Assessment Survey of Agricultural Household (70th round), conducted by the NSSO during 2013. The data-set provides information on the income, sources of income, productive assets, resource availability, awareness, education, access to modern technology, agricultural practices and other important aspects in the field of agriculture and related occupations. As no other data provides information on the income of agricultural households by source² of occupation, it is the most important data-set to understand the pattern of non-farm diversification in rural India. The survey has taken a sample of 2,084 respondents and 727 households from the states of Bihar and Punjab considered to be fit for the macro-level analysis. Apart from this data, the article also makes use of Employment and Unemployment Surveys since 1983 till the latest Periodic Labour Force Survey (PLFS) 2017–2018, to understand the trends and the growth pattern of the employment situation in the two selected states.

Methods

There are numerous methods to define a household as a non-farm household. It can be based on its principal source of income or the time spent on non-farm activity. However, restricting oneself to such definitional scopes may lead to a loss of information about the non-farm activity conducted by agricultural households. Therefore in the present study, we have considered every household which earns income from any non-farm source during a year. We call these households ‘mix-income households’. Those who are earning only from the agriculture sector have been referred to as ‘agricultural households’. To ascertain the determinants of rural non-farm diversification, a binary logistic regression model has been applied. The model has been chosen given the nature of the dependent variable as a binary outcome variable.

Occupational Shift in This Millennium

The share of employment in the agricultural sector has declined sharply over the last two decades in India in general and in Punjab, in particular. This employment shift has been observed across Indian states, irrespective of their levels of economic development. For instance, the share of agriculture in rural Bihar has declined by 32 percentage points during 1999–2000 to 2017–2018 as compared to less than three percentage points between 1983 to 1999–2000 (Figure 1). Similarly, for Punjab this share has declined by 32 percentage points as compared to 10 percentage points during the previous period. It can further be observed from Figure 1 that a change in employment composition has happened gradually in Punjab as the industrial and services sector started gaining share in total employment in the late 1990s. However, in the case of Bihar this transformation is visible during the mid-2000s. First, these trends show that the rural areas were witnessing a significant employment shift away from agriculture during the last two decades as compared to the previous decades. Second, the shift has been observed for both agriculturally advanced states as well as backward ones. Lastly, this transformation, which was more prominent in Punjab even during 1983 to 1999–2000, is now significant for Bihar too.

Figure 1.

Source: Produced from Employment and Unemployment Surveys (from 1983 to 2011–2012) and Periodic Labour Force Survey (PLFS, 2017–2018), Government of India.

The transformation of rural employment can be scrutinised further with regard to the growth of employment (Figure 2). As reflected in Figure 2 , employment growth during the last two decades has been negative in the agriculture sector for both states. It is the rural industrial and services sectors which are growing in the recent period. As far as industrial sector is concerned, it is the construction sector, especially in Bihar, which grew at a high rate of growth (11.6 per cent per annum) during 1999–2000 to 2017–2018. The employment growth scenario in Punjab and Bihar differs in the form of a changing sectoral composition. However, the overall pattern suggests that the employment shift has taken away substantially from agriculture during this new millennium.

Figure 2.

Source: Produced from Employment and Unemployment Surveys (from 1983 to 2011–2012) and Periodic Labour Force Survey (PLFS, 2017–2018), Government of India.

In sum, both states are witnessing a shift of workforce away from agriculture. This poses the question: Is this shift bringing any positive employment outcomes? To understand this question, it is important to examine who actually accesses rural nonfarm employment.

Mapping Income Sources of Rural Households

To understand the type of rural household engaged in non-farm activities see Figure 3 which presents the percentage of households engaged in each occupation. It can be observed that the share of households engaged in cultivation increases with land size which is obvious for both the states. However, it is noteworthy that households having no land or limited land are largely dependent upon wage/salaried employment. Also, 90 per cent of landless households in Bihar and 80 per cent in Punjab are engaged in rural wage labour. Similarly, households belonging to semi-marginal categories are highly dependent on wage labour in both states. Non-agricultural enterprises seem to be a limited option for households in both states and across land size.

Figure 3.

Source: Produced from unit-level data of Situational Assessment Survey of Agricultural Households, National Sample Survey Office (NSSO), 70th round, Government of India.

Further, it is important to understand the primary source of income of the rural households. It has been observed that wage/salaried work or non-farm enterprise as a primary source of income is inversely related to the size of land. It can be seen from Figure 4 that households which are landless or with smaller categories of land sizes have reported wages and salaries as their primary source of income. This share is particularly higher for Punjab. It can be seen that all households belonging to large category of land size have reported cultivation as their primary source of income.

It is to be noted that the share of non-farm enterprises does not constitute a significant share across all the land classes (Figure 4). In the non-farm sector, wage labour is the main driver of employment growth in both states. This raises the question of whether the opportunities in the non-farm sector are as remunerative as they are in the farm sector. To understand these questions, let us examine the level of income from each source.

Figure 4.

Source: Produced from unit-level data of Situational Assessment Survey of Agricultural Households, National Sample Survey Office (NSSO), 70th round, Government of India.

Table 1 provides the level of income by source for various land size groups. It can be observed that the mean income from non-farm enterprises is lower than any other sources across various land classes. It also needs to be mentioned that it constitutes only 4.4 per cent and 6.5 per cent of the total income of the rural households in Punjab and Bihar respectively. For the landless, semi-marginal and marginal, wage employment constitutes a high proportion as the main source of income followed by animal husbandry and cultivation. These categories in both states thus seem to be diversified.

Table 1.

Average Monthly Income of Rural Households from Various Sources by Land Possession (in rupees)

Size Class of Land Possessed	Punjab					Bihar
Size Class of Land Possessed	Crop Cultivation	Animal Husbandry	Wage/Salaries	Non-farm Enterprises	Total Income	Crop Cultivation	Animal Husbandry	Wage/Salaries	Non-farm Enterprises	Total Income
Landless	–	–	5,094 (91.5)	470 (8.5)	5,143 (100.0)	10 (0.2)	630 (13.8)	3,781 (82.5)	160 (3.5)	4,582 (100.0)
Semi-marginal	318 (3.7)	730 (8.4)	6,600 (76.4)	991 (11.5)	8,639 (100.0)	549 (18.4)	345 (11.5)	1,883 (62.9)	215 (7.2)	2,991 (100.0)
Marginal	7,061 (49.4)	3,045 (21.3)	3,645 (25.5)	555 (3.9)	14,306 (100.0)	1,884 (56.9)	503 (15.2)	829 (25.1)	93 (2.8)	3,308 (100.0)
Small	12,128 (72.4)	1,382 (8.3)	2,430 (14.5)	800 (4.8)	16,741 (100.0)	3,553 (80.9)	– (0.0)	818 (16.1)	715 (14.1)	4,390 (100.0)
Semi-medium	21,513 (74.7)	2,014 (7.0)	4,522 (15.7)	766 (2.7)	28,815 (100.0)	7,143 (62.7)	805 (7.1)	2,961 (26.0)	476 (4.2)	11,385 (100.0)
Medium	36,179 (83.7)	2,902 (6.7)	3,552 (8.2)	570 (1.3)	43,203 (100.0)	26,013 (98.8)	– (0.0)	380 (1.4)	103 (0.4)	26,319 (100.0)
Large	75,813 (85.0)	10,151 (11.4)	1,262 (1.4)	2,015 (2.3)	89,241 (100.0)	35,033 (100.0)	– (0.0)	0.0 (0.0)	0.0 (0.0)	34,485 (100.0)
All Sizes	10,845 (60.0)	1,662 (9.2)	4,774 (26.4)	800 (4.4)	18,082 (100.0)	1,722 (45.8)	307 (8.2)	1,485 (39.5)	245 (6.5)	3,759 (100.0)

Source: Computed from unit-level data of Situational Assessment Survey of Agricultural Households, National Sample Survey Office (NSSO), 70th round, Government of India.

Note: Figures in parenthesis are percentage income from each source (Landless = < 0.01, semi-marginal = 0.01–0.40, marginal = 0.41–1.00, small = 1.01–2.00, semi-medium = 2.01–4.00, medium = 4.01–10.00, large = 10.00+).

On the contrary, it can be observed that households with medium and large land categories in Punjab also earn a sizeable proportion of their income from non-farm sources whereas in Bihar these categories are entirely dependent on cultivation. It indicates that there exist some remunerative non-farm opportunities in Punjab in which large land category households can engage which is missing in the case of Bihar. Secondly, in both states, smaller land category households are engaged in rural non-farm activities where rural households do not face any barrier to entry. However, in adopting lucrative options, they may face high barriers to entry such as capital, higher education, skills and social cohesion. Therefore, the chances of adopting better non-farm opportunities may be unequal among rural households. The analysis in the succeeding section explores these determinants which are probably responsible in facilitating the choice of non-farm diversification.

Determinants of Household Participation in Rural Non-farm Employment

The attempt here is to provide a brief discussion on the various drivers of non-farm employment and how these influence a household’s decision to participate in it.

Table 2 provides the descriptive summary of variables considered for the analysis. The dependent variable in the analysis is income sources. This variable is modified with a binary outcome as ‘only agricultural and allied activities’ and ‘mixed income sources’ (agriculture and allied, along with non-farm activities). Households which are performing only agriculture-related activities are considered ‘only agricultural household’ and ‘mixed occupation household’ are those considered for the second case. It is found in Table 2 . that the proportion of mixed income households is higher in Punjab (68 per cent) than that of Bihar (52 per cent ).

As household characteristics are important factors, the analysis has taken variables such as the gender of the household head, the educational level of the head, social group and age. The hypothesis is that the association of a household with a particular social background makes a difference to the probability for adopting non-farm opportunity. In Punjab, the percentage of the forward caste group is higher in the mixed occupation category (87.7 per cent). However, it is the other way around in Bihar. It indicates that the forward caste group has more access to non-farm income sources in Punjab while in Bihar they are less dependent on this source of income. This pattern is statistically significant³ for both the states. The education of the household head is considered a factor which facilitates decision-making as well as entry in the rural non-farm sector. Where the percentage of a better-educated household head is more for a mixed occupation category in Punjab, it is lower in Bihar. It indicates that the households with better education do not prefer to engage in rural non-farm employment activities in Bihar.

Gender also plays a major role in diversifying the income of a household. The share of male family head is very high for both categories of occupation ( Table 2 ). It indicates that male members have better access to not only agriculture but also to non-farm activities in both the states. Age is not a significant factor for both states as the difference in the average age is less between the two categories. It has been found that the average size of a household is relatively higher in the mixed occupation category ( Table 2 ). It has been consistently argued in the literature studied that households with more members are also highly diversified in terms of non-farm employment.

Table 2.

Descriptive Analysis of Variables for Punjab and Bihar

	Punjab			Bihar
	Agricultural Occupation (N = 285)	Mixed Occupation (N = 439)	t-stat/ Chi-Square	Agricultural Occupation (N = 985)	Mixed Occupation (N = 1,075)	t-stat/ Chi-Square
Dependent Var. (%)	39.3	60.7	–	47.8	52.2	–
Dichotomous Independent Variables
Gender of Head (male %)	95.4	91.9	3.54**	97.9	89.8	56.91***
Edu-level of Head (Above Primary %)	53.8	60.4	2.98*	53.3	44.9	14.54***
Caste (Forward %)	60.4	87.7	62.92***	45.4	41.7	3.86**
Landownership (less than 1ha)	24.6	54.4	63.57***	48.4	70.5	105.59***
House Structure (Pucca %)	99.6	98.4	2.42^ns	90.7	85.4	13.8***
Continuous Independent Variables
Age of the Head (years)	52.3	54.2	–1.84*	50.6	50.0	1.08^ns
Household Size (numbers)	5.0	5.6	–3.30***	6.0	6.3	–2.44**
Monthly Per-capita Household Farm Income (₹)	5670.0	2793.5	5.70***	1097.4	517.8	8.33***

Source: Produced from unit-level data of Situational Assessment Survey of Agricultural Households, National Sample Survey Office (NSSO), 70th round, Government of India.

Notes: ***indicates the level of significance at 1 per cent, ** at 5 per cent and * at 1 per cent, ^ns stands for not significant.

It can be observed that households with no land or with small land holdings are engaged more in non-farm employment than those households having more land. This pattern is evident in Bihar where 70 per cent of the households with no land or up to one hectare of land are engaged in mixed income occupations ( Table 2 ). Further, households with pucca (permanent) structures participate more in agriculture in Bihar indicating that katcha (semi-permanent) structured households are more diversified. For Punjab, the comparison is not possible between katcha and pucca households as this variable is not statistically significant. The fact that the average monthly per-capita income from agriculture is lower for the mixed occupation category indicates that households with better farm income do not diversify their sources of income in the non-farm sector.

Determining the Participation of Households in the Rural Non-farm Sector

The determinants have been examined to ascertain the dominant forces which facilitate non-farm diversification for rural households. Therefore, binary logistic regression has been employed to analyse these factors. Binary logistics is preferred as the dependent variable has binary outcomes: first, whether a household is earning only from agriculture and second, if it has a mixed income source (agriculture as well as non-farm sources). The detail of coding for each variable is presented in Appendix 1 . The model has been estimated for both the states separately to have a comparative assesment of underlying factors.

Among the other household factors, it is found that households with male members are more likely to participate in a non-farm sector in Bihar, indicating that the non-farm sector is biased against gender ( Table 3 ). This finding is in line with earlier studies claiming that the female workforce face barriers to entry in non-farm activities. However, the analysis does not find any conclusive evidence for this phenomenon in the case of Punjab. Further, as the age of the household head increases, there is a higher probability to diversify into non-farm activities in both states. It may be expected that as the age of the household head increases, liabilities and number of dependents also increase. Therefore, to enhance the levels of income, which could not be possible within the agriculture sector for resource deficient households, a household decides to participate in the rural non-farm sector.

Table 3.

Estimates of the Logistic Regression Analysis

Dependent Variable = Non-farm Participation (Agriculture = 0, Agriculture + Non-farm = 1)
Explanatory Variables	Punjab (N = 724)		Bihar (N = 2,060)
Explanatory Variables	Odds ratio	Std. error	Odds ratio	Std. error
Household Variables
Gender Head (male = 1)	1.71^ns	0.365	5.34***	0.255
Age Head	1.02*	0.007	1.006*	0.004
Size of Household	1.14**	0.043	1.05***	0.016
Social Group (forward = 1)	3.33***	0.232	1.01^ns	0.116
Education Head (above primary = 1)	1.05**	0.026	1.03*	0.014
Landownership: (Reference = small (1 to 2 ha))
Landless and Marginal (0 to 1ha)	2.62**	0.020	2.04**	0.293
Semi-medium and Medium (2 to 10 ha)	0.07^ns	0.303	0.45^ns	0.290
Large (more than 10 ha)	1.88*	0.084	0.75^ns	0.290
House Structure (pucca = 1)	4.84^ns	1.122	1.34**	0.153
Farm Income	1.03^ns	0.554	0.20^ns	0.320
Model Information
Log Likelihood	835.9		2627.4
Cox & Snell	0.17		0.103
Nagelkerke R	0.230		0.138
Hosmer and Lemeshow Test Chi-Square	10.67^ns		6.71^ns

Source: Produced from unit-level data of Situational Assessment Survey of Agricultural Households, National Sample Survey Office (NSSO), 70th round, Government of India.

Notes: *** indicates the level of significance at 1 per cent, ** at 5 per cent and * at 10 per cent, ^ns stands for not significant

Further, as the size of the family increases, a rural household seeks income from non-farm sources as well. These factors support the arguments for distress-driven, non-farm diversification for both the states. In addition, forward caste households are more likely to participate in the non-farm sector in Punjab. However, the results are not significant for Bihar. Education is one of the main factors that is considered a barrier against entering the non-farm sector which is significantly noticeable for both the states. The results indicate that those who are educated till the primary-level and above are more likely to diversify their income sources towards non-farm sectors.

The pattern between the size of land ownership and participation in the non-farm sector is not linear in Punjab. Those who are landless or own marginal land are likely to participate in non-farm employment; however, this relation is insignificant for those who have medium land holdings. Further, households with large land sizes are also likely to participate in non-farm activities, but only in Punjab. In Bihar, there are only landless and marginal categories that are more likely to participate or diversify income from non-farm sources. This suggests that there are possibly only lowly remunerative non-farm opportunities available in Bihar. Therefore, households having more access to land do not participate in this sector. On the contrary in Punjab, there also exists a small number of relatively remunerative non-farm opportunities where households with large land holdings participate to further enhance their income.

The type of structure taken as proxy of household conditions does not show a significant result in Punjab as there are pratically no katcha houses left in the state. However in Bihar, households with pucca structures are more likely to participate in the rural non-farm labour market which indicates that there are some barriers to their entry into non-farm sectors such as minimum credit availability or capital available to the family. Farm income can play an important role in generating non-farm opportunities as surplus from the farm sector may support a non-farm venture. However, in both states, farm income does not show any significant impact on the household’s participation in the non-farm sector.

Conclusion

The analysis was carried out to understand the changing employment composition in rural Punjab and Bihar separately. In Punjab, the rural non-farm sector gained importance during the 1980s when the agricultural sector growth generated new employment opportunities whereas it remained an important source of livelihood for landless households. But in the new millennium, when agriculture is facing huge distress and has become unviable for small farming communities, especially in Punjab, the non-farm sector has become the last resort for those who are moving out.

It has been found that a large number of less remunerative non-farm opportunities have emerged in both states. It means a distress-driven diversification in the non-farm sector is more dominant over a demand-driven employment growth in this sector. It is also evident from this analysis that households with a large family size participate more in this sector which indicates that households with more dependent members lead to a lower asset base. It forces them to enhance income levels through participating in the non-farm sector. Complementing this assertion, we also found that households without any land or less land participate in these activities to enhance their level of income. There seem to be a few non-farm employment opportunities available in Punjab in which large farm households also participate whereas in Bihar such a pattern is missing. Therefore the size of land is a major determining factor of participation in both states.

Barriers to entry in this sector take different shapes in both states. Where caste is a major barrier to entry in Punjab, gender is a dominant factor in Bihar. Caste plays an important role as social relations and linkages facilitate in gaining remunerative employment opportunities. In Bihar, social stigma against gender restrains female workers from joining non-farm opportunities. Whereas studies have pointed out that male workers often opt out for work in the non-farm sector and females take care of agricultural work. The level of education seems to be very important in facilitating households moving away from agriculture.

There are two types of policy challenges needed to address the issue of employment transformation in rural India; first, making the non-farm sector more lucrative for the rural masses through public policy provisioning and favouring this sector; second, providing lucrative, non-farm employment opportunities through skill development, providing cheap credit, and socially inclusive policies. Thus, this article argues for first, creating clear institutional ownership over rural non-farm matters which should not be treated as parking lot for the unemployed; second, investing in sustainable rural financial systems that can fill capital deficiency; third, concerted efforts in developing skills; and lastly, serious policy attention towards creating an institutional infrastructure to make non-farm activities accessible and viable to all.

Footnotes

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

Appendix

Appendix 1.

Description of the Variables Included in the Analysis

Variables	Indicator	Description
Non-farm Participation	Agriculture = 0, Mixed = 1	Dependent Variable: Participation Only in Agriculture and Related Activities; and Participation in Agriculture as well as Non-agricultural Activities
Gender	Male = 1, Female = 0	Gender of the Household Head
Age	Continuous Variable	(in Years)
Size of Household	Continuous Variable	(in number)
Social Group	Lower Caste = 0, Upper Caste = 1	Lower Includes SC, ST and OBC Together
Education above Primary Level	Below Primary = 0, Above Primary = 1	–
Landownership	Landless and Marginal–1, Small–2, Semi-medium–3 and Medium + Large–4	Landless (0 to 1 ha), Small (1 to 2 ha), Semi-medium and Medium (2 to 10 ha), Large (more than 4 ha)
House Structure	Katcha–0, Pucca–1	Pucca Includes Semi-pucca also
Monthly Per-capita Household Farm Income	Continuous Variable	in Rupees

Notes

References

Abraham

(2013). Missing labour or consistent ‘de-Feminisation?’ Economic & Political Weekly, 48(31) 99–108.

Asfaw

, Simane

, Hassen

, & Bantider

(2017). Determinants of non-farm livelihood diversification: Evidence from rainfed-dependent smallholder farmers in north central Ethiopia (Woleka sub-basin). Development Studies Research, 4(1), 22–36.

Barrett

C. B.

, Reardon

, & Webb

(2001). Nonfarm income diversification and household livelihood strategies in rural Africa: Concepts, dynamics, and policy implications. Food Policy, 26(4), 315–331.

Bhalla

(1993). Patterns of employment generations in India. The Indian Journal of Labour Economics, 36(4), 504–527.

Bhaumik

S. K.

(2007). Growth and composition of rural non-farm employment in India in the era of economic reforms. Indian Economic Journal, 55(3), 40–65.

Binswanger-Mkhize

H. P.

(2012). India 1960-2010: Structural change, the rural non-farm sector, and the prospects for agriculture. Department of Agricultural and Resource Economics, University of California. http://are.berkeley.edu/documents/seminar/Binswanger.pdf

Centre for Budget and Governance Accountability. (CBGA) (2020). Decoding the priorities: An analysis of Union Budget 2020–2021. http://www.cbgaindia.org/wpcontent/uploads/2020/02/Decoding-the-Priorities-An-Analysis-of-Union-Budget-2020-21-1.pdf

Chadha

G. K.

(2003). The rural nonfarm sector in the Indian economy: Growth, challenges, and future direction. In Gulati

& Fan

(Eds.), The dragon and the elephant (pp. 343–364). IFPRI, Johns Hopkins University Press.

Ellis

(1998). Household strategies and rural livelihood diversification. The Journal of Development Studies, 35(1), 1–38.

10.

Haggblade

, Hazell

P. B.

, & Reardon

(2009). Transforming the rural nonfarm economy: Opportunities and threats in the developing world. Johns Hopkins University Press.

11.

Haque

(2016, 19 November. Challenges of accelerated and inclusive agricultural growth in India [Lecture delivered]. 10th Dr Verghese Kurien Foundation Day, KIIT University Bhubaneswar. https://archive.ksrm.ac.in/wp-content/uploads/2016/11/10th-Dr-Kurien-Foundation-Lecture-Nov-19th-2016.pdf

12.

Harriss-White

(2014). Labour and petty production. Development and Change, 45(5), 981–1000.

13.

Himanshu (2019, 29 April. India’s farm crisis: Decades old and with deep roots. The India Forum. https://www.theindiaforum.in/article/farm-crisis-runs-deep-higher-msps-and-cash-handouts-are-not-enough

14.

Jatav

, & Sen

(2013). Drivers of non-farm employment in rural India. Economic & Political Weekly, 48(26–27), 14–21.

15.

Jha

(1997). Agricultural labour in India. Vikas Publishers.

16.

Jha

(2016). Labour in contemporary India. Oxford University Press.

17.

Lanjouw

, & Shariff

(2004). Rural non-farm employment in India: Access, incomes and poverty impact. Economic & Political Weekly, 39(40), 4429–4446.

18.

Seng

(2015). The effects of nonfarm activities on farm households’ food consumption in rural Cambodia. Development Studies Research, 2(1), 77–89.

19.

Srivastava

, & Srivastava

(2010). Women, work and employment outcomes in rural India. Economic & Political Weekly, 45(28), 49–63.

20.

Tacoli

(1998). Rural-urban interactions: A guide to the literature. Environment and urbanization, 10(1), 147–166.

21.

Unni

(1994). Inter-regional variations in non-agricultural employment in rural India: An exploratory analysis. In Visaria

& Basant

(Eds.), Non-agricultural employment in India: Trends and prospects (pp. 289–329). SAGE Publications.

22.

Vaidyanathan

(1986). Labour use in rural India: A study of spatial and temporal variations. Economic & Political Weekly, 21(52), A130–A146.

23.

Visaria

, & Basant

(Eds.) (1994). Non-agricultural employment in India: Trends and prospects. SAGE Publications.