saumyavb72@yahoo.co.in) is with the (anirban_kundu2006@rediffmail.com i
SPECIAL ARTICLEmarch 21, 2009 vol xliv no 12 EPW Economic & Political Weekly701972-73 to 1999-2000 the total number of workers (usual status, primary and secondary status) and the share of employment in this non-farm sector had increased from 352.4 lakh to 726.4 lakh and 14.3% to 23.8%, respectively (Table 3). It is generally argued that the cause of such expansion of the RNF sector in India isthe particular pattern of growth of agriculture. Consequently, farm-non-farm linkages have become an important issue of discussion. 3 Review of Literature on Farm-Non-farm LinkageAs our basic issue of discussion is the impact of crop-diversifica-tion and agricultural land-conversion on the dynamics of the RNF sector, the literature on farm-non-farm linkages may provide us with the appropriate building blocks for the subsequent analysis. Hence, we take up a journey through the relevant literature. Vaidyanathan (1986) in his study showed that where agriculture is unable to provide suffi-cient employment theRNF sector picks up a part of the slack. Thus, an inverse relationship between farm and non-farm employment was proposed (see also Bhaumik 2002). However, contrary to this “residual sector” hypothesis there is a vast literature that proposes a posi-tive relationship between farm growth and expansion of RNF activities. Hazell and Haggblade (1990) consid-ering state and district level data for India calculated that on an average a Rs 100 increase in agricultural earnings is associated with a Rs 64 increase inRNF income (see also,IFPRI 1985; Hazell et al 1991). It was found by Haggblade et al (1989) that marginal consumption of locally produced non-foods is large (about 35%) in Asia (India and Malaysia). Mellor (1976) hypothesised that the green revolution generated increased demand for locally produced labour-intensiveRNF goods and services. But the asser-tion that even the big farmer class could be the driving force for non-farm growth has been questioned by several researchers (Dunham 1991).Using examples from India, Pakistan and Taiwan, Johnston and Kilby (1975) highlight the importance of production linkage. They point to small farmers’ demand for fertilisers and others similar inputs as well as that for equipment along with repair services provided by local artisans. According to Hazell and Roell (1983) small- and middle-sized peasant farmers have much higher propensity to consume labour-intensive rurally produced goods compared to that of their larger counterparts. In such cases non-farm employment multipliers are also substantial (Krishna 1976). Data from Punjab raises similar doubts about the significance of large farmer-based agriculture. In this respect we can site the study by Bhalla and Chadha (1983) on Punjab, where they found Table 1: Growth Rates of Agricultural Output and Employment (in %) 1985-86 to 1994-95 1995-86 to 2002-03Output growth rate 3.4 1.8 1983-931993-99 Employment growth rate 2.6 0Source:D Kapur’s comment on“Pre- and Post-Reform India”by S Bhalla and T Das,in India Policy Forum 2005-06,Vol 2, Sage,India,2006.Table 2: Employment Share of Different Sectors (in %) Sector 1983 1993-94 1999-2000Primary 6965.560.4Secondary 13.8 14.8 16.8Tertiary 17.2 19.7 22.7Source:Indian Economy by R Dutt and K Sundaram,S Chand Publication.Table 3: Number (in Lakh) and Percentage of RNF Workers (to total workers) (us - ps+ss basis)1972-73 1977-7819831987-781993-941999-2000352.4 364.9 459.2 559.6631.8 726.414.3 16.518.421.721.623.8Source: “Employment Diversification in Rural India: A State Level Analysis” by S K Bhaumik in Indian Journal of Labour Economics, Vol 45(4), 2002 (p 721).agriculture has become an important issue of research. Substan-tial empirical work has been done on the question of farm-non-farm linkages in India. It is proposed that farm sector growth in-duces expansion of non-farm sector through demand (consump-tion) and supply side (forward and backward) linkages. Expan-sion of non-farm economy, in its turn, helps agriculture to grow further through demand boost as well as through supply side support (Mellor 1976; Ranis and Stewart 1993). Thus, a growth multiplier process operates involving both farm and non-farm sectors.2 However, such a mechanism is argued to be dependent largely on the condition of egalitarian growth in agriculture (Harriss 1991). Agricultural growth based on equitable asset dis-tribution is supposed to raise demand for locally produced labour-intensive goods and services through forward as well as backward linkages.Nevertheless, at present Indian agri-culture is experiencing significant re-source reallocations. Two such funda-mental changes are crop-diversification and agricultural land-conversion for industrialisation. The government of India is also shifting its policy focus from basic cereals production to production of so-called high value “non-food” com-mercial crops, especially vegetables, fruits and flowers. It is argued that such crop-diversification would boost em-ployment and income generation. Moreover, it is supposed to ensure sustainable cultivation (Joshi et al 2004). On the other hand, huge tracts of agricultural land are being converted for construction of industrial zones leading to loss of basic agricul-tural production. Such changes may, however, have profound impacts not only on the basic food crop (cereals) economy but also on the RNF sector. We, in this paper, try to discuss the probable impacts of crop-diversification and agricultural land-conversion on the vastRNF economy engaged in petty production.The paper is constructed as follows. Section 2 presents some stylised facts on the employment situation in India. Section 3 provides the review of literature on farm-non-farm linkages. InSection 4, we note a few empirical observations on farm-non-farm relations pertaining to some states of India. In Sections 5 and 6 we try to explore the probable impacts of crop-diversification and agricultural land-conversion on the RNF sector. Section 7 concludes. 2 Some Stylised Facts on the Employment SituationBefore going into the core of the issues let us present certain information on the overall employment situation in India. In Table 1, the growth rates of agricultural output and employment are shown. It is pretty clear that the situation is alarming. It calls for a serious search for effective alternatives. So far as the secondary and tertiary sectors are concerned there is also not much hope for employment generation (refer Table 2).This is obvious given the nature of technology used in these sectors. In this situation, the RNF sector is said to provide some respite. Such optimism is also supported by the fact that from 
SPECIAL ARTICLEEconomic & Political Weekly EPW march 21, 2009 vol xliv no 1271that middle peasantry (5-12.5 acre) and not the big or very big farmers is the dominant source of demand for non-farm goods and services. A survey on Arni, Tamil Nadu (Harriss 1991) has revealed that agricultural expansion has not encouraged local production of labour-intensive consumer goods; rather it has disproportionately expanded trade in “big city” products. Based on this survey Harriss concludes that not only agricultural output and income growth but also its pattern has serious implications for the growth of labour-intensive RNF sector.3 In particular he asserts that it is the egalitarian growth process in agriculture that is essential for the development of decentralised labour-intensive non-farm sector.Similarly, the importance of redistributive land reforms in pro-moting localised RNF activities has been emphasised by several researchers (Saith 1991 and 1992; Ray 1994). In fact, small farm-based agriculture is supposed to help the growth of the RNF sector through both the demand and supply side linkages. Agri-cultural expansion generates demand for local inputs especially rudimentary tools and farm services and for locally produced cheap consumer goods. On the other hand, such an expansion ensures the vital local supply of food and other raw materials.4 In this context, the observations of the West Bengal Human Development Report 2004 need special attention. It is argued that as the growth of agricultural output and income in West Bengal during the 1980s and 1990s has been achieved on the founda-tions of land reforms, it has created a wider market for mass consumption goods and for different locally produced inputs including farm-related services such as petty processing and transportation, petty trading of crops, etc. On the other hand, it has also ensured the vital micro (household or village level) food security. Thus, the newly evolved class of small and marginal farmers and the bargadars concentrating mainly on basic crops production have been the source of food and raw materials for the local non-farm sector. Consequently, a mutually beneficial (reinforcing) endogenous growth process develops. Summaris-ing, we can say that the small farm economy specialising in non-mechanised production of basic crops generated on the one hand substantial agricultural employment and hence large market for cheap mass consumption goods, and on the other, demand for locally produced and supplied inputs of production and farm services. Consequently, a large section of the rural population gets engaged in local non-farm activities. According to this report, 40% of the total rural workers in the state are engaged in such non-agricultural activities. A very large part of these activities are, in fact, undertaken in household enterprises. The National Sample Survey Organisation (NSSO 2001) survey on the informal sector in India shows that the proportion (per 1,000) of enterprises located within household premises is very high (571) for West Bengal and it is well above the national average (448). This is particularly plausible given the food security of the households ensured through formal access to the produce itself.4 Empirical Observations on Farm-Non-farm LinkageIn support of our argument that the labour-intensive RNF sector is much more integrated with a small farm-based agriculture concentrating on primary commodities or basic cereals pro-duction, we go for a simple empirical exercise. Using data from NSSO (2005) on productive assets for farm and non-farm business possessed by the farmer households we have con-structed Table 4 for three representative states and for all India average.We have selected purposively three states, West Bengal, Tamil Nadu and Punjab. Agriculture in these three states had performed well in the 1980s and early 1990s. However, Punjab’s progress was based on the growth of traditional foodgrain pro-duction (particularly wheat), while Tamil Nadu’s agriculture was one of the most diversified. Contrarily, in West Bengal agriculture was small farm based though concentrating mainly on rice cultivation. From Table 4 it can be inferred that marginal (up to 1 hectare) and small (1.01 to 2 hectares) farmers are much more tied up with non-farm sector across India as well as in all the three representative states. These farmers largely depend on such equipment for farm and non-farm businesses, which are generally produced in the local labour intensive non-farm sector. Primary implements like sickle, plough, axe, etc, are mostly owned by the marginal and small farmers compared to medium (2.01 to 4 hectares) and large (4.01 to 10 hectares) ones. On the other hand, capital-intensive implements like power tiller, tractor, pump are used proportionately more by upper class farmers. Similarly, the machinery and equipment meant for non-farm business are mainly used by the small and marginal farmers. This is true for India as well as for the three states. Another interesting result is that this inclination is maxi-mum in West Bengal followed by Punjab and Tamil Nadu. This observation again supports the view that extensive employ-ment diversification as experienced in rural West Bengal is largely due to its particular pattern of land and agricultural income distribution.A similar kind of inference could also be drawn if average monthly (net) income from non-farm business per farmer household for the three states is compared. This is done in Table 4: Average (Per Hectare) No of Productive Assets for Farm and Non-Farm Business Possessed (Per 1,000 Farmer Households by Size Class of Land Possessed)(season: kharif) Size Class (in hectares) Up to 2 2.01-4 4.01-10 Up to 2 2.01-4 4.01-10 Up to 2 2.01-4 4.01-10 Up to 2 2.01-4 4.01-10Items West Bengal Tamil Nadu Punjab All India[For farm business] (primary implements) Sickle,chaff-cutter,axe, spade,chopper,plough 5,064 2,707 1,426 3,700 2,363 1,175 5,543 3,402 1,906 5,776 3,038 1,564Pump 101 120 91 242 279 192 360 400 220 128 141 91(Capital intensive implements) Power tiller,tractor 23 20 26 5 13 11 84 243 162 16 35 26[For non-farm business] Machinery and equipments 158 12 0 90 1 14 99 81 1 140 22 10Source: Calculated from NSS 59th round, Report No 497 (59/33/5), pp A28, A31, A35, A37.
Punjab West Bengal Tamil Nadu 
SPECIAL ARTICLEEconomic & Political Weekly EPW march 21, 2009 vol xliv no 1273in collection, transportation, storage, processing, packaging and finally retailing could effectively realise the potentials of crop-diversification (Rao et al 2006; Sen and Raju 2006).7 Hence, the intrusion of big capital in this elaborate chain of activities becomes imminent either through spot contracts or through the integrated institution of contract farming. The diversifiedHVC is siphoned off for “big city” consumption or for exports and on the other hand, modern inputs and modern farm services are intro-duced into the agricultural sector. The same channel could even be used to sell “big city” products in rural areas. Three clear con-sequences of this whole process could be: (i) lack of access to local (basic) food supply making the non-farm sector vulnerable (Sen and Raju 2006); (ii) shifting of demand for agricultural inputs and farm services from local non-farm sector to the modern industry (Basant 1994); and (iii) “big city” cheap products displacing non-farm consumer goods (Harriss 1991).Thus, the non-farm sector faces challenge from the introduc-tion of modern industrial commodities. The insertion of capital displaces the non-farm population, only a part of which could be internalised into the agricultural modernisation process. In this context, we can refer to the study of Basant (1994) on Gujarat. It is noted that “… growth of land productivity (or per capita out-put) based mainly on cropping pattern changes may not generate adequate impulses for the development of rural non-agricultural activities. It may be partly because the processing of output is likely to be concentrated in urban areas and the techniques of production may be capital-intensive”. Thus, the pattern of agri-cultural growth in Gujarat is such that it could not boost signifi-cantly theRNF activities of “rural located-rural linked” type. Rather, “urban linked” (though rurally located) firms played a significant role and whateverRNF activities developed was capital-intensive in nature (ibid).Agricultural diversification may also jeopardise macro, micro and especially, local and household food security.8 It could really be a problem for the petty producers both in agriculture produc-ingHVC and shifting away from basic cereals and in non-farm, getting displaced due to agricultural resource reallocation. Food insecurity at local as well as at household levels makes these farmer households dependent on external sources. However, as the non-farm households lose their work they are doubly affected. They neither have the access to local food supply nor do they have sufficient entitlement to exchange with.9It is evident from Table 5 (p 72) that crop-diversification is bound to affect the food security of low-income earning farmer households who spend a large part of their income on basic food items such as cereals, pulses, etc.This argument is supported by Table 6 as well. We have con-structed the average monthly per capita consumption expendi-ture for farmer households across different size class of land pos-sessed for the three representative states (as before) and for all-India. It shows that except Tamil Nadu, for all other cases the households with relatively smaller amount of land spend signifi-cantly higher proportion of their income on cereals and pulses compared to the bigger counterparts. Moreover, for West Bengal, Punjab and all-India averages the ratios of share of expenditure on cereals, grams, pulses and their products to that on vegetables and fruits and to that on beverages, processed food, etc, fall dras-tically as the amount of land possessed rises. Hence, the wide-spread practice of crop-diversification with the motto of “ever-green revolution” under the guidance of the National Farmers Commission or National Horticulture Commission may lead to a significant erosion of micro food security, which, in turn, may destabilise the indigenous rural non-farm sector. All the probable outcomes discussed above have resemblance with the Hymer-Resnick (1969) result where expansion of export-able cash crop sector of the colony displaces the rural traditional non-farm activities. In our present context, Ranis and Stewart’s (1993) “unfavourable post-colonial archetype” could also be referred to as a similar case. Ranis and Stewart, however, show that Taiwan’s agricultural progress based on “non-traditional” crop helped the development of “rural industry”, mostly processing and other related activities. But they categorically mention that such a linkage could be established because of the deliberate policy favouring the RNF sector (“Z-goods sector”). On the con-trary, the Philippine macro policy encouraged modern industry (“U-industry”), which led to the displacement of Z-goods sector even if, like Taiwan, the Philippines also concentrated on export of non-traditional processed crops. The Indian case of diversification under the macro framework of corporate contract farming seems to be closer to the case of Philippines rather than that of Taiwan. Table 6: Monthly Per Capita Consumer Expenditure SharesOver Different ‘Food’ Items for Farmer Households Across Size Class of Land PossessedSize Class of Land Per Capita MPCE Share of Expenditure Share of Share of Possessed Consumption Class on Cereals, Cereal Expenditure Expenditure on Expenditure Substitutes, Grams, on Vegetables, Beverages, (in Rs) Pulses and Their Fresh and Refreshments and Products Dry Fruits Processed Food (%) (%) (%)West Bengal < 0.01 435.47 420 – 470 48.83 13.65 4.210.01-0.40 437.73 420 – 470 48.83 13.65 4.210.41-1 561.13 525 – 615 44.95 14.44 4.331.01-2 731.50 615 – 775 41.26 14.17 5.152.01-4 896.98 775 – 950 39.59 13.94 5.714.01-10 987.54950 + 35.41 15.76 6.47Tamil Nadu < 0.01 456.36 420 – 470 41.95 14.13 9.500.01-0.40 507.50 470 – 525 41.76 13.91 9.550.41-1 583.18 525 – 615 39.07 14.68 11.071.01-2 724.31 615 – 775 39.38 14.26 11.272.01-4 838.86 775 – 950 38.10 13.93 10.874.01-10 862.04 775 – 950 38.10 13.93 10.87Punjab < 0.01 624.13 615 – 775 25.39 11.26 6.960.01-0.40 658.44 615 – 775 25.39 11.26 6.960.41-1 833.44 775 – 950 22.36 11.66 7.081.01-2 940 775 – 950 22.36 11.66 7.082.01-4 1010.34 950 + 19.34 11.97 7.284.01-10 1351.03 950 + 19.34 11.97 7.28All-India < 0.01 417.63 380 – 420 47.01 13.50 4.650.01-0.40 434.54 420 – 470 44.65 13.43 5.020.41-1 485.81 470 – 525 42.92 13.36 5.181.01 -2 572.36 525 – 615 39.94 13.68 5.662.01-4 670.00 615 – 775 36.33 13.38 6.174.01-10 841.09 775 – 950 32.69 13.31 6.45Source: Computed from NSS 59th Round, Report No 495 (59/33/4), pp 10, A75, A78, A82, A84; and Report No 497 (59/33/5), pp A187, A189, A191.
SPECIAL ARTICLEmarch 21, 2009 vol xliv no 12 EPW Economic & Political Weekly74One may argue, as crop pattern changes RNF sector survives on thisHVC cultivation itself. Evidently, it would not be possible since small farms would be tied up with the big agro-based multi-national corporations (MNCs) through contract farming – diversi-fication, processing, retailing chain both for inputs and market. Even the small and marginal farmers for sheer survival could accept this arrangement as crop-diversification becomes unviable without such contracting. This whole agreement between the small and marginal farmers and big agri-business firms would drive out petty non-agricultural activities by squeezing the demand as well as the supply side channels. Though the huge population engaged in different types of RNF activities gets expropriated, only a minor fraction of it would be able to attach itself with the so-called modernisation process involving formal sector and HVC based agriculture. 6 Land-Conversion and the RNF SectorThere are many studies which confirm that the access to land is a crucial factor explaining access to rural non-agricultural activi-ties/employment opportunities (West Bengal Human Develop-ment Report 2004; Ray 1994; Saith 1991). Thus rural households owning small plots of land or sharing landownership (or culti-vation) rights could diversify to different other localised petty activities over and above basic agriculture, using the benefits of agricultural growth itself. Micro food security ensured through access to land played an important role for rural employment diversification. In fact, such diversification enabled the house-holds to neutralise the seasonal fluctuations of agricultural pro-duction. On the other hand, job diversification by the farmer households reduced the pressure on land as agriculture is man-aged more efficiently. Thus, small farm-based agriculture on the one hand, and employment diversification by these farmer house-holds, on the other, sustain a mutually reinforcing endogenous growth process.Nevertheless, the present debate hovering around the issue of agricultural land conversion for industrialisation is not paying sufficient attention to the probable impacts on thisRNF sector. Agricultural land-conversion may seriously affect the micro food security shaking the very foundation of non-farm growth. The expropriation of peasantry displaces the localised labour inten-sive non-farm activities because the demand as well as supply side linkages between farm and non-farm sectors get snapped.On the other hand, as the formal sectors expand, demand for HVC rather than that of basic cereals increases. Expansion of modern industry and services increases the demand for skilled labour and thereby disproportionately raises the income of the upper stratum of the population. This changes the demand pat-tern in favour of HVC shifting away from basic cereals. Conse-quently, the prices of HVCs rise creating incentive for the farmers to shift away from basic crop production. This may create not only micro food insecurity but also shortage of raw materials choking off the non-farm sector. Thus, the contractionary effects of agricultural land-conversion on the RNF sector get reinforced as the so-called modern diversified agriculture becomes inte-grated with the “advanced” sectors severing its close links with the labour intensiveRNF economy.Thus, land-conversion coupled with crop-diversification can doubly damage the foundations of the RNF sector. While, land alienation disturbs micro food security, compensating the loss of food production due to conversion through crop-diversification reduces demand for non-farm inputs and local non-farm services. However, while, the issue of compensation in the context of land acquisition is intensely debated, there is almost nothing for the people engaged in localised non-farm sector. 7 ConclusionsA very important proposition that emerges from the preceding analysis is that sectoral policies such as crop-diversification or agricultural land-conversion for industrialisation should not be formulated in isolation. Such policies should particularly take into account the corresponding impacts on the rural labour- intensive non-farm sector which is accepted as a dynamic segment of the economy having enormous employment generation poten-tials. The policies of crop-diversification to raise farm income and land-conversion for rapid industrial progress should form only parts of a more comprehensive broader project encompass-ing all the major sectors of the economy.Crop diversification could be a socially beneficial policy when it is complemented by extensive infrastructural facilities, finan-cial and technological support, etc, especially for the localised micro (labour-intensive) enterprises engaged in processing, storing, grading and packaging. Furthermore, specific support to organise the marketing network is very crucial for a policy of crop-diversification targeted towards small farmers. Only such a comprehensive policy package can boost non-farm production and service activities involving indigenous resources and utilising local labour. An appropriate institutional set-up should be organ-ised so that a labour-intensive growth multiplier process is trig-gered off involving both the diversified agriculture and neigh-bouring non-farm sector. At present, processing is mainly done by small-scale non-farm units (99.4%) where 86.8% of total processing sector employment is generated (Dev and Rao 2005). Only an appropriate institutional arrangement (eg, network of cooperatives engaged in production as well as distribution) along with government support for technological upgradation, infra-structure, farm-services, etc, are essential. However, land-conversion may have a serious impact and needs detailed planning to counterbalance the adverse effects. Notes 1 It has been found (Bhaumik 2002) in the Indian context that the RNF sector can play effective role in reducing rural poverty. 2 In fact, growth of farm and non-farm sectors are not separate phenomena, rather parts of an integrated endogenous process (Sanyal 2007).3 De Janvry and Sadoulet (1993) in the context of Latin America observe that be-cause of highly skewed distribution of land and farm income the benefits of agri-cultural growth could not be appropriated by the RNF sector in the absence of significant consumption linkages. In this context, we should refer to the Indian data on distribution of landholding which could provide some explanation for the existence of a large RNF sector. From the Agricultural Statistics at a Glance 2003 (Government of India) it could be found that the share of small and marginal op-erational holdings with size up to two hectares increased from 70% to more than 80% during the period 1970-71 to 1995-96. However, the corresponding share for the large holdings of four hectares and above declined from 14% to almost 7% during the same period. This typical pattern of landholding may have contributed to the existence of the vast non-farm sector in India. 4 Ranis and Stewart (1993) also indicate at such linkages in the case of Taiwan as compared to the Philippines. 
