You are here: Home » Opinion » Columns
Business Standard

Raghav Gaiha & Vani S Kulkarni: Non-farm employment and poverty reduction

Raghav Gaiha & Vani S Kulkarni  |  New Delhi 

Infrastructure helps reduce poverty directly and indirectly through the expansion of non-agricultural activities.
Poverty estimates from the 61st round of the National Sample Survey (NSS) conducted in 2004-05 have revived the debate on whether accelerated growth in recent years has trickled down to the rural poor. The trickle-down mechanisms usually take the form of lower food prices and higher wage rates.
Since the 61st round is fully comparable to the 50th round of the NSS, the poverty rates are comparable over the period 1993-94 to 2004-05. A recent analysis of the (grouped) data from the 61st round shows that the head-count ratio fell from 37.2 per cent in 1993-94 to 28.7 per cent in 2004-05. The pace of poverty reduction, however, was lower than in the previous two decades. While the annual reduction was ""0.88 per cent during 1983-93, it slowed down to ""0.77 per cent during 1993-04. A supplementary analysis based on comparable Employment-Unemployment surveys of the 55th round (1999-00) and the 61st round of the NSS (2004-05) points to a sharp reduction in the head-count ratio in rural India (that is, from 34 per cent in 1999 to about 25 per cent in 2004""an annual reduction of ""1.8 per cent). On some assumptions, the conclusion is drawn that the bulk of the poverty reduction between 1993-2004 occurred during 1999-04. This is somewhat intriguing as the growth of agriculture, wage employment and wages decelerated during the latter period. A spatial configuration of poverty offers interesting clues:
Distribution of self and wage employment in
non-agricultural activities in rural India, 2004
employment in
activities (%)
employment in agriculture (%)
activities (%)
agriculture (%)
BIMARU 46.51 22.53 11.56 16.71
(Bihar, Madhya Pradesh,Rajasthan and Uttar Pradesh)
South 50.41 27.44 13.02 9.41
(Andhra Pradesh, Karnataka, Kerala, Tamil Nadu, Maharashtra and Gujarat)
East 46.88 21.81 15.77 15.33
(Assam, West Bengal and Orissa)
Total 48.82 18.68 12.67 17.46
Source: Gaiha and Imai (2007) Row-wise shares add to 100.

  • The growth of casual real wages decelerated at the all-India level""for male workers from 2.80 per cent per annum during 1993-1999 to 1.38 per cent during 1999-2004""but accelerated in most of the poor states (for example, Assam, Madhya Pradesh and Orissa).

  • During 1993-99, the Consumer Price Index for Agricultural Labourers (Food) or CPIAL (Food) rose annually at 8 per cent. After 1999-2004, the rate of the growth of food prices to that of non-food prices dropped to 31 per cent during 1999-2004, with the lowest ratio in Assam, Bihar, Karnataka and Madhya Pradesh. As these are the states where the bulk of the poor are located, it is not surprising that they recorded the highest reduction in poverty (Himanshu, 2007, "Recent Trends in Poverty and Inequality: Some Preliminary Results", Economic and Political Weekly, February 10).

  • Some other significant features of the period 1999-2004 are that both non-agricultural wage and self-employment accelerated. While non-agricultural wage employment growth rose from 2.68 per cent per annum during 1993-99 to 3.79 per cent during 1999-04, the self-employment grew at a faster rate (from 2.34 per cent annually to 5.72 per cent). A recent analysis (Gaiha, Raghav and Katsushi Imai (2007) "Non-Agricultural Employment and Poverty in India""An Analysis Based on the 60th Round of the NSS", Cambridge: MA: Harvard Centre for Population and Development Studies, draft) throws new light on the poverty reducing potential of non-farm activities.
  • Much of the development literature has focused on production and consumption linkages between farm and non-farm activities. The dynamics of agricultural and non-agricultural linkages are driven by the primacy of agriculture in the development process. While there is considerable merit in this argument, it overlooks the crucial role of infrastructure in reducing poverty directly as well as through the expansion of non-agricultural activities.
    As noted earlier, participation in non-agricultural activities takes two forms, viz, wage and self-employment. The distribution of these categories by region, computed from the 60th round of the NSS, is given in the Table.
    As may be noted, the shares of both wage and self-employment in non-agricultural activities in total regional rural employment are the lowest in the so-called BIMARU states but the gaps between them and the southern states are relatively small. So, if some of the poorest states have performed better in poverty reduction in recent years, this could be due to the rapid expansion of non-agricultural activities.
    Some econometric evidence suggests the following.

  • The probability of being poor varies inversely with education, as also with technical education.

  • Landless households are more likely to be poor.

  • Both SC and ST households are more likely to be poor, as also those belonging to OBCs.

  • The probability of being poor declines with participation in non-agricultural wage and self-employment.

  • States with better infrastructure support (such as road length, telephone density, electricity) exhibit lower poverty, as also greater non-farm employment opportunities.
  • Within the non-agricultural sector, participation in sub-sectors (food processing, manufacturing, trading and other non-agricultural activities) also varies with household characteristics. For example, women are less likely to participate in food processing, manufacturing and trading activities; while both school and technical education enable participation in all sub-sectors, technical education favours participation in manufacturing more; SC/STs and OBCs are less likely to participate in any of these sub-sectors, relative to others; however, better infrastructure facilitates expansion of all these activities.
    Two policy concerns must be addressed: one is that targeted interventions may be unavoidable to ensure that disadvantaged groups such as SC/ST/OBCs have easier access to non-farm employment opportunities to overcome persistent poverty.
    The second and often overlooked concern is that the absorption of surplus rural labour in non-farm activities is conditional on rapid expansion of school and technical education and better infrastructure.
    So infrastructure""including school and technical education, roads, communication networks, electricity supply""helps reduce poverty directly as well as indirectly through the expansion of non-agricultural activities.
    While the budgetary outlays for infrastructure in 2007-08 reflect this concern, their financing remains a worry.
    Raghav Gaiha and Vani S Kulkarni are at Harvard's Centre for Population and Development Studies

    First Published: Thu, April 26 2007. 00:00 IST