ISBN 92-9066-474-6 Order code BOE 035 614–2004

ICRISAT-Patancheru(Headquarters)Patancheru 502 324Andhra Pradesh, IndiaTel +91 40 23296161Fax +91 40 23241239icrisat@cgiar.org

ICRISAT-Niamey(Regional hub WCA)BP 12404Niamey, Niger (Via Paris)Tel +227 722529, 722725Fax +227 734329icrisatsc@cgiar.org

ICRISAT-Nairobi(Regional hub ESA)PO Box 39063, Nairobi, KenyaTel +254 20 524555Fax +254 20 524001icrisat-nairobi@cgiar.org

ICRISAT-BulawayoMatopos Research StationPO Box 776,Bulawayo, ZimbabweTel +263 83 8311-15Fax +263 83 8253/8307icrisatzw@cgiar.org

ICRISAT-BamakoBP 320Bamako, MaliTel +223 2223375Fax +223 2228683icrisat-w-mali@cgiar.org

ICRISAT-Maputoc/o INIA, Av. das FPLM No 2698Caixa Postal 1906Maputo, MozambiqueTel +258-1-461657Fax +258-1-461581icrisatmoz@panintra.com

ICRISAT-LilongweChitedze Agricultural Research StationPO Box 1096Lilongwe, MalawiTel +265-1-707297/071/067/057Fax +265-1-707298icrisat-malawi@cgiar.org

The International Crops Research Institute for the Semi-Arid Tropics (ICRISAT) is a non-profit, non-political, international organization for science-based agricultural development. ICRISAT conductsresearch on sorghum, pearl millet, chickpea, pigeonpea and groundnut – crops that support the livelihoodsof the poorest of the poor in the semi-arid tropics encompassing 48 countries. ICRISAT also sharesinformation and knowledge through capacity building, publications, and information and communicationtechnologies (ICTs). Established in 1972, it is one of 15 Centers supported by the Consultative Group onInternational Agricultural Research (CGIAR).

Contact information

About ICRISAT

Visit us at www.icrisat.org

Overco

min

g P

overty in

Ru

ral Ind

ia

Citation: Rao KPC, Bantilan MCS, Singh K, Subrahmanyam S, Deshingkar P, Rao P Parthasarathy and Shiferaw B.2005. Overcoming Poverty in Rural India: Focus on Rainfed Semi-Arid Tropics. Patancheru 502 324 AndhraPradesh, India: International Crops Research Institute for the Semi-Arid Tropics. 96 pp. ISBN 92-9066-474-6.Order code BOE 035.

Abstract

This study is a comprehensive analysis of the nature, the dynamics and the determinants of poverty in thesemi-arid tropics (SAT). It uses the sustainable livelihood framework to identify the underlying determinantsof poverty and the pathways in moving out of poverty. Quantitative analysis based on the most recent NationalSample Surveys (NSS) of India (1999–2000 round) generated classifications of relevant typologies includingagroecological zones, occupational groups, irrigation access and social groups. It shows that poverty in India isconcentrated most in the SAT and humid areas. It identifies caste and occupational groups facing highestincidence of poverty. The typologies systematized the establishment of linkages between rural poverty and thesemi-arid tropics.

The study presents a synthesis of evidence and lessons learnt from the village level studies (VLS) by theInternational Crops Research Institute for the Semi-Arid Tropics (ICRISAT) and other household level casestudies – income and expenditure levels, incidence of poverty and under-nutrition and coping strategies whenfaced with droughts and other income shocks. The empirical analysis uses both the quantitative and thequalitative data to document the existing and emerging evidences on the spatial and the social dimensions ofpoverty. The poverty diagnostics and strategic assessment of the issues confronting SAT generated acomprehensive perspective from both the micro and the macro levels, thus deepening our understanding ofthe multi-dimensional nature of SAT poverty.

The study reviews the existing poverty alleviation programs (their merits and demerits and successes andfailures) and concludes with a set of recommendations on strategies to increase the contribution of SATagriculture to poverty reduction.

Résumé

La présente étude fait une analyse détaillée de la nature, de la dynamique et des déterminants de la pauvretédans les zones tropicales semi-arides (SAT). Elle se s’appuie sur le cadre de moyens de subsistance durablespour identifier les facteurs qui déterminent la pauvreté et explorer les pistes qui permettent d’en sortir. Uneanalyse quantitative, basée sur les enquêtes les plus récentes faites en Inde (1999-2000), à partir d’échantillonsnationaux (National Sample Surveys – NSS), s’est traduite par des classifications pertinentes - zones agro-écologiques, groupes professionnels, accès à l’irrigation et groupes sociaux. Cette analyse montre, qu’en Inde,la pauvreté se concentre surtout dans les SAT et les zones humides. Elle permet d’identifier les castes et lesgroupes professionnels qui subissent le plus les conséquences de la pauvreté. Les typologies ont systématisél’établissement de liens entre la pauvreté en milieu rural et les zones tropicales semi-arides.

L’article résume les preuves et les enseignements issus des études menées au niveau villageois (VLS) parl’Institut international de recherche sur les cultures des zones tropicales semi-arides (ICRISAT) et d’autresétudes de cas réalisées au niveau des ménages – niveaux des revenus et des dépenses; incidence de la pauvreté,et sous-alimentation et stratégies d’adaptation en cas de sécheresse et autres contrecoups liés au revenu.L’analyse empirique s’appuie à la fois sur les données quantitatives et qualitatives en vue d’étayer les preuvesexistantes et émergentes, relatives aux dimensions spatiales et sociales de la pauvreté. Le diagnostic de lapauvreté et l’évaluation stratégique des problèmes que connaissent les zones tropicales semi-arides, ont permisd’avoir une perspective générale, d’un point de vue à la fois micro et macro, renforçant ainsi notre compréhensionde la nature multidimensionnelle de la pauvreté dans les SAT.

L’article examine les programmes qui existent en matière d’atténuation de la pauvreté (avantages/inconvénients;succès et échecs) et se termine par une série de recommandations sur les stratégies qui permettent d’améliorerla contribution de l’agriculture dans les SAT à la réduction de la pauvreté

Overcoming Poverty in Rural India:Focus on Rainfed Semi-Arid Tropics

ICRISATInternational Crops Research Institute for the Semi-Arid Tropics

Patancheru 502 324, Andhra Pradesh, India

IFADInternational Fund for Agricultural Development

Via del Serafico, 107, 00142 Rome, Italy

2005

KPC Rao, MCS Bantilan, Katar Singh, S Subrahmanyam,Priya Deshingkar, P Parthasarathy Rao and Bekele Shiferaw

About the authorsKPC Rao Principal Scientist (Village Level Studies), Global Theme on SAT Futures and

Development Pathways, International Crops Research Institute for the Semi-AridTropics, Patancheru 502 324, Andhra Pradesh, India.

MCS Bantilan Global Theme Leader – SAT Futures and Development Pathways, InternationalCrops Research Institute for the Semi-Arid Tropics, Patancheru 502 324, AndhraPradesh, India.

Katar Singh Former Director, Institute of Rural Management, Anand 388 001, Gujarat, India.

S Subrahmanyam Senior Fellow, Centre for Economic and Social Studies, Hyderabad, Andhra Pradesh,India.

Priya Deshingkar Research Director, Andhra Pradesh Livelihood Options Project, OverseasDevelopment Institute, UK.

P Parthasarathy Rao Senior Scientist (Economics), Global Theme on SAT Futures and DevelopmentPathways, International Crops Research Institute for the Semi-Arid Tropics,Patancheru 502 324, Andhra Pradesh, India.

Bekele Shiferaw Scientist-Resource Economics, Global Theme on SAT Futures and DevelopmentPathways, International Crops Research Institute for the Semi-Arid Tropics.

© 2005 by the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT)

All rights reserved. Except for quotations of short passages for the purpose of criticism and review, no part of this publication may be

reproduced, stored in retrieval systems, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or

otherwise, without prior permission from ICRISAT. It is hoped that this copyright declaration will not diminish the bona fide use of

research findings in agricultural research and development.

The opinions expressed in this publication are those of the authors and not necessarily those of ICRISAT. The designations employed

and the presentation of material in this publication do not imply the expression of any opinion whatsoever on the part of ICRISAT

concerning the legal status of any country, territory, city or area, or concerning the delimitation of its frontiers or boundaries. Where

trade names are used, this does not constitute endorsement of or discrimination against any product by the Institute.

AcknowledgmentsWe express our thanks to the International Fund for Agricultural Development (IFAD) for inviting ICRISATto contribute a background paper for developing its Country Strategic Opportunities Paper (COSOP). Wewould like to express special thanks to CH Hanumantha Rao and MV Nadkarni for their valuable commentsto further articulate the message of this paper. We also wish to express thanks to Pravesh Sharma, ShyamKhadka and Julian Hamilton-Peach for their useful comments. We acknowledge the assistance of the Ministryof Human Resource Development, Government of India, for providing the NSS data in time for the analysis.

We would also like to thank the members of the working team – PK Mehta, R Padmaja, GV Anupama,Dharmendra Kar, Gloria Christina and B Ravindra – for their valuable assistance.

Our thanks to NVN Chari, H Padmini and VN Krishnan for their administrative and secretarial assistance.

Contents

Executive summary .................................................................................................................................... 1

1. Introduction and background ................................................................................................................. 5

1.1 Genesis and background .................................................................................................................. 5

1.2 Importance of SAT and SAT agriculture in India’s economy............................................................. 6

1.3 Important interventions in SAT agriculture and their impacts .......................................................... 7

1.4 Objectives of the study and research questions ................................................................................ 7

1.5 The sample and sources of data ....................................................................................................... 8

1.6 Structure of report .......................................................................................................................... 8

2. Sustainable livelihoods approach and analytical framework .................................................................... 9

2.1 Introduction .................................................................................................................................... 9

2.2 Sustainable livelihoods approach and framework ............................................................................. 9

2.3 A new sustainable livelihoods framework for SAT agriculture ........................................................ 10

2.4 Multi-dimensional view of poverty ................................................................................................. 11

2.5 Complementary framework for identifying determinants of poverty .............................................. 13

3. Levels and determinants of rural poverty in the semi-arid tropics of India ........................................... 14

3.1 Introduction .................................................................................................................................. 14

3.2 Sources of data and methodology .................................................................................................. 14

3.3 Where are the poor concentrated? ................................................................................................ 15

3.4 Who are the rural poor? ................................................................................................................ 16

3.5 Why are they poor? ....................................................................................................................... 18

3.6 Key factors influencing poverty and crop productivity ................................................................... 21

3.7 Major findings related to SAT ........................................................................................................ 22

3.8 Synthesis of results and policy implications .................................................................................... 22

4. Major livelihood strategies and current interventions in SAT agriculture .............................................. 24

4.1 Introduction .................................................................................................................................. 24

4.2 Livelihood portfolios and changes over time................................................................................... 24

4.3 Coping strategies adopted by farmers and agricultural laborers ...................................................... 28

4.4 Current interventions in SAT agriculture and their impacts ........................................................... 30

4.5 Concluding remarks ....................................................................................................................... 31

5. Emerging changes and issues in SAT agriculture ................................................................................... 33

5.1 Introduction .................................................................................................................................. 33

5.2 Biophysical problems ..................................................................................................................... 33

5.3 Technological issues ....................................................................................................................... 34

5.4 Economic problems ....................................................................................................................... 35

5.5 Institutional/organizational problems ............................................................................................. 35

5.6 Administrative and management problems ..................................................................................... 37

5.7 Public policy-related problems ....................................................................................................... 37

5.8 Concluding remarks ....................................................................................................................... 39

6. A development strategy for SAT agriculture ........................................................................................ 40

6.1 Introduction .................................................................................................................................. 40

6.2 Priority thematic areas for possible development interventions ...................................................... 40

6.3 Priority geographical areas for possible interventions ...................................................................... 45

6.4 Priority areas for future research .................................................................................................... 46

6.5 Building partnerships and linkages .................................................................................................. 48

6.6 Capacity building and development ............................................................................................... 48

6.7 Concluding remarks ....................................................................................................................... 48

Bibliography ............................................................................................................................................ 50

Appendixes .............................................................................................................................................. 55

Appendix 1. The semi-arid tropics and other agro-ecological zones ...................................................... 57

Appendix 2. Evolution and current thinking of the sustainablelivelihoods approach ............................................................................................................................ 65

Appendix 3. Determinants of poverty ................................................................................................. 67

Appendix 4. Identifying determinants of income poverty – framework ................................................ 83

Appendix 5. Methodology for computing poverty indicators ................................................................ 85

Appendix 6. A list of selected centrally-sponsored rural developmentprograms currently underway in SAT, India .......................................................................................... 88

Appendix 7. A brief critique of some major programs and lessons learned ........................................... 89

1

Executive Summary

Understanding the underlying determinants ofpoverty and the pathways to its alleviation is a majorchallenge. To start with, the Sustainable LivelihoodsApproach (SLA) emphasizes on the rural poor sector,particularly in marginalized areas, to empower, value andwork on what matters to them. This ‘pro-poor peoplecentered’ perspective is essential for the development ofan impact-driven policy and investment strategy. In linewith this perspective, the research team appeals to theprinciples of sustainable livelihoods in developing anoverarching framework for the entire study. The SLAwas updated and expanded to capture the essentialdeterminants and dynamics of poverty in the process ofidentifying primary constraints and opportunities inSAT. In particular, it added emphasis on the aspects ofmarket access, agricultural research and technology andpower relations as important determinants of povertyand drivers of change in SAT. It recognizes the multi-dimensional nature of poverty, encompassing economic,social and governance perspectives.

The empirical analysis used both quantitative andqualitative data to document the existing andemerging evidences on the spatial and socialdimensions of poverty. Quantitative analysis wasbased on the most recent National Sample Surveys(NSSs) of India (1999–2000 round) which hasgenerated various measures of poverty. Classificationsby relevant typologies including agro-ecologicalregions, occupational groups, access to irrigation,social groups, among others, systematized theestablishment of the linkages between rural povertyand SAT in the context of India. District, state andnational level estimates based on India’s NSS of70,000 sample households were generated to comparethe incidence of poverty across regions. The resultsunderlined the economic dimensions of poverty basedon income, consumption expenditures, wages,productivity and other measurable parameters.Interesting trends showed the varying levels ofincidence of poverty among occupational sectors aswell as agro-ecological zones and other relevantgroupings. It also brought out implications regardingthe dependence of the poor on agriculture in the SATregion.

The major findings generated from the quantitativeassessments are as follows:• The total number of poor in rural India is estimated

at 147.5 million during 1999-2000, out of which81.0% or 119.9 million poor are concentrated inthe humid and the SAT regions.

This report is a strategic assessment of rural poverty inthe semi-arid tropics (SAT) involving a comprehensiveanalysis of the dynamics and determinants of poverty,and the major issues of the SAT agriculture in India.The objectives of the strategic assessment are to• establish a linkage between the SAT agriculture and

rural poverty;• understand the critical issues of SAT agriculture in

India;• recommend strategies to increase the contributions

of SAT agriculture in reducing poverty; and• recommend priority interventions and research

directions.Ultimately, the analyses of major issues confronting

the SAT agriculture and the alternative strategies andrecommendations are intended to serve as a basis foridentifying priority areas for future investments thatwould increase the contribution of SAT agriculture inreducing poverty in a sustainable manner.

An assumption underlying this strategic study isthat the current interventions in the SAT agricultureare inadequate and have to be enhanced andintensified. In fact, the SAT agriculture explicitly boreunfavorable policy and institutional neglect, leading toinadequate investments. It is expected that thelivelihoods of the rural poor in the SAT can beenhanced by addressing the key constraints of the ruralsector and the major issues confronting the SATagriculture. It is also expected that the strategyprepared as an output of this study will result inchanges in the priorities of the government and donorsto hasten the contribution of SAT agriculture inreducing poverty in a sustainable manner.

The SAT region is a vital and important sector onwhich a large number of poor in India depend on. Anoverview of the state of SAT agriculture in India gives aperspective on the nature of poverty and the dynamicsof the external environment surrounding it. Thisoverview presents a background of the SAT in terms ofcoverage, population, major crops and its importanceto India’s economy. The major issues affecting the SATagriculture involve several dimensions: social, technical,structural, policy, institutional, human resources, andbiotic and abiotic constraints. The developments inSAT reflects the pervasiveness of poverty (predomi-nantly a rural phenomenon), growing constraints onwater, continuing concerns about malnutrition, changesin production and consumption patterns, and effectsof government policies and further economicliberalization on the competitiveness of dryland crops.

2

Overcoming Poverty in Rural India

• The humid and the SAT regions have roughly thesame size of India’s rural population and theytogether account for 72% of the rural population.The semi-arid temperate zone accounts for 21%and the arid zone accounts for about 2% of the ruralpopulation.

• The SAT and humid zones have high incidence ofpoverty (about 24%) and the arid and semi-aridtemperate zones have low incidence of poverty(12.6% and 14.6%). By and large, all areas with lowirrigation have the highest incidence of poverty inall the zones.

• The incidence of poverty is found to be the highestamong the Scheduled Tribes (ST) at 39.6%followed by the Scheduled Castes (SC) at 28.56%in contrast to 16.4% among general population.Scheduled Tribes are concentrated in the humidand SAT zones, which partly is the reason for highincidence in these two regions. The low irrigatedareas in the humid and SAT zones have a highconcentration of these social groups. Thecorrelation coefficient between poverty incidenceand Scheduled Tribes is found to be high at 0.47 forrural India.

• Among five occupational groups, the incidence ofpoverty is the highest among agricultural laborers(wage workers) at 35.5% and the lowest among thecultivators at 13.9%. The non-agricultural laborershave significantly lower poverty levels than theagricultural laborers.

• Incidence of poverty among agricultural laborers isuniformly high in all the zones and it is distinctlyhigher in the low irrigated areas.

• The correlation between poverty and theproportion of agricultural laborers is positive andvery high in all the zones except in the high-irrigated areas of the SAT.

• The correlation between poverty incidence and theproportion of non-agricultural laborers is negativeand statistically significant indicating that thedevelopment of non-farm activities is an importantpolicy intervention to reduce rural poverty.

• Agricultural wages and cereal prices are importantdeterminants of rural poverty. High agriculturalwages tend to decrease the incidence of povertywhile high cereal prices tend to increase theincidence of poverty.

• Semi-arid temperate and humid zones areagriculturally developed but have high populationpressure on land. On the other hand, the SAT zoneis backward in agriculture but has moderatepressure on land. Only the arid zone has very lowpressure on land and in spite of low agricultural

productivity, productivity per capita is quite high.Hence, poverty levels tend to be low.

• Except in the semi-arid temperate zone, consider-able sections of non-poor were also found to bepurchasing rice and wheat from the PublicDistribution System (PDS). This implies a leakagein the system that needs to be rectified.

• The rural poor from the humid zones had a highercoverage under the anti-poverty programs thanthose in other zones. Relatively low utilization ofthe anti-poverty programs by the rural poor in theSAT and arid zones highlights the critical issue ofconstrained access by these bypassed groups.

• Arid and semi-arid temperate zones have markedlyhigher levels of standard of living whereas thehumid and SAT zones are having relatively lowlevels of standard of living.

• Agricultural development, higher wages andemployment opportunities contribute to povertyreduction. The effect of wages is much strongerthan that of productivity growth mainly becausemany of the rural poor are net buyers.

• The semi-arid temperate zone has well developedmarkets whereas the SAT zone is lagging in marketinfrastructure.The qualitative analysis substantiated other equally

important dimensions not covered in the quantitativeanalysis above. The report presents a synthesis ofevidences and lessons learned from the ICRISAT’sVillage Level Studies (VLS) and other household levelcase studies. In particular, it provided empiricalevidence about the vulnerability of the poor to varioussources of risks and shocks as well as the capacity toaccess physical, financial and social resources andnetworks in the risky environments of SAT. Theanalysis captured welfare indicators involving the levelof human development and the extent of vulnerabilityand insecurity among individuals or households. Themajor livelihood strategies adopted by the populationin the SAT region are elucidated with particularemphasis on the livelihood portfolios of the rural poor.It is on this basis that the list of major issues that needto be addressed to strengthen livelihoods in the SATregions are identified.

Qualitative evidences drawn from the VLS, casestudies and other micro-level data highlight thefollowing:• There have been many changes in the patterns and

sources of livelihood in the SAT region over the last25 years or so in response to changes in technology,infrastructure, markets, urbanization and com-mercialization of agriculture. Diversification ofsources of income and shifts in cropping patterns

3

Executive summary

and food habits are occurring in most of the SATareas in India.

• Real incomes of the VLS sample households morethan doubled over the last 25 years, 1975–78 to2001–02. And the structure of income underwentremarkable changes.

• The share of income from crops declined in theSAT villages except in villages in the more reliablerainfall regions and those which had access tosupplemental irrigation. The share of income fromlivestock showed a marginal decline. The share ofearnings from agricultural wages also declinedconsiderably, while that of earnings from non-farmactivities increased sharply.

• Diversification of income sources has increasedincluding new high value commodities andmigration.

• About one-third of the sample households werebelow the poverty line (with less than Rs 20,000net annual household incomes) and the proportionof the poor estimated from micro-level studies waseven higher than what was reported by theNational Sample Surveys Organization (NSSO) forthe relevant districts.

• The consumption expenditure data pointed to aconsiderable undernutrition among the SAThouseholds relative to other regions.

• Besides low incomes, the SAT households face ahigh probability of drought-occurrence andconsequent shortfalls in incomes. Nearly one half ofthe sample households adopted coping strategiessuch as borrowing, reduced consumption,participation in labor market, sale of assets andmigration to survive the droughts.On the whole, at micro and macro levels, the

poverty diagnostics and strategic assessment of theissues confronting the SAT generated a comprehensiveperspective, which enhanced our understanding of themulti-dimensional nature and dynamics of poverty,including the linkages between rural poverty and itsdeterminants and the opportunities for future priorityinvestments. This understanding is vital in (1)identifying and designing appropriate interventions tohelp the poor achieve preferred livelihood outcomes,(2) improving the poor’s access to productive naturalresources and technology, and (3) evolving strategiesfrom an investment angle and enhance the standardsof living of the poor.

The report concludes with a set of recommenda-tions on strategies to increase the contribution of theSAT agriculture resulting in the reduction of poverty.Major questions such as what is the future ofagriculture in less favorable regions such as SAT; whatkind of agriculture would be sustainable; what kind of

policy, research and extension support is needed tomake agriculture financially viable for the majority offarmers; what priority areas for future investmentswould increase the contribution of SAT agriculture inreducing poverty in a sustainable manner; havecontributed to the development of a strategy for theSAT agriculture. Future research strategy inagriculture should be responsive to the felt needs ofthe poor farmers in spite of the complex relationshipbetween agricultural research and poverty reduction;and aim at generating yield-increasing, drought-resistant and labor-intensive technologies suitable foragro-climatic conditions.

The priority thematic areas for possible develop-ment intervention in the SAT region are• improved access to supplementary irrigation along

with complementary inputs (techniques forharvesting and storage of rain water, recharging ofgroundwater aquifers and regulation ofgroundwater extraction, adoption of watershedmanagement approach, rationalization of pricing ofwater and power reflecting their opportunity costs,improving water use efficiency of irrigatedagriculture, specification and enforcement ofproperty rights in water);

• wider diffusion of new dryland technologiesadapted to local conditions;

• reclamation and productive use of wastelands,fallow lands and common pool lands;

• diversification of agriculture with a higher priorityto livestock development;

• marketing and commercial orientation ofagriculture;

• revamped and targeted public distribution system;• strengthening of support systems and basic

infrastructure;• better-targeted wage employment schemes to

address transient poverty;• institutional innovations (securing women’s

participation, demand-driven agricultural extensionsystem; using mass media to create awarenessabout the problems of SAT agriculture); and

• re-orienting public policies (protecting the interestsof migrant labor, ensuring food security throughprovision of wage-paid employment, and rationaliz-ing subsidies on agricultural inputs).The low irrigated SAT areas covering Maharashtra,

Madhya Pradesh, Gujarat, Karnataka, AndhraPradesh, Bihar and Rajasthan; the medium irrigatedareas covering parts of Rajasthan, Madhya Pradesh,Gujarat, Karnataka, Andhra Pradesh and Bihar; andthe high irrigated areas of SAT spanning over TamilNadu, Andhra Pradesh, Bihar, and Gujarat are thepriority geographical areas for intervention.

4

Overcoming Poverty in Rural India

In setting priorities for research, the involvement oftarget users in the process is important. Participatorypriority setting can lead to different diagnoses andrecommendations for future agricultural research andproject implementation. A synthesis of participatorysurveys cited in this report has shown how farmers’own perceptions and priorities regarding constraints toproduction and the quality of agricultural serviceswent beyond just resource endowments and allocation(climatic and environmental factors, access to land,labor, skills, and draft power) to the quality of publicand private agricultural services. This process helpedto identify the livelihood diversification strategiesincluding changing farming practices, more reliance onnon-farm sources of income, and modified patterns ofexchange and consumption. Consequently, the studyrecommends the following priorities for research:• developing cost-effective methods and techniques

of integrated watershed development and manage-ment;

• techno-economic assessment of water saving microirrigation technologies and determinants of theiradoption in SAT;

• breeding of water-efficient drought-tolerantvarieties of rainfed crops; shift in focus from cropimprovement of sorghum and millets as food grainsto their improvement as feed grains;

• markets and policy; and alternative options formaking the typical farm family financially viable

through optimum mix of farm and non-farmenterprises;

• value addition and better utilization of dryland crops;• mainstreaming of gender in the SAT agricultural

development strategies;• role of power relations and social exclusion in

designing and implementing effective interventionstrategies;

• economics of postharvest technologies in SAT;• institutional innovations for reducing the high

transaction costs and marketing costs; and• harnessing modern information and communica-

tion technologies for outreach and technologydissemination as well as interfacing with farmers,policy makers, Non-Governmental Organizations(NGOs) and research agencies.The report begins with an overview of the state of

SAT agriculture in India and an analytical framework forthe study based on essential principles of sustainablelivelihood. Linkages between rural poverty and the SATagriculture are identified and major livelihood strategiesand current interventions in the SAT agriculture arepresented in subsequent units of the report. The majorissues affecting the SAT agriculture in India, includingbiophysical, technical, economic and institutional issuesare discussed. The report concludes with a set ofrecommendations on strategies to increase thecontribution of SAT agriculture in reducing ruralpoverty in a sustainable manner.

5

1. Introduction and Background

point for the analysis and understanding of povertyand designing of interventions in the SAT agriculture.

1.1 Genesis and backgroundThe strategy of the International Fund for AgriculturalDevelopment (IFAD) for Asia and the Pacificenvisages a focus on the less-favored areas. In focusingon this niche area, the strategy intends to ensurecomplementarity with the poverty-reductioninitiatives of governments, particularly for thelandless, and with those of the World Bank and otherdonors in areas such as health, education andinfrastructure. The following are fundamental to theregional strategy to reduce rural poverty:

• the changing of unequal gender relations to increasewomen’s ownership and control of assets, and theireffective participation in community managementaffairs;

• enhancing the productivity of staple food in less-favored areas;

• reforming property and tenurial rights of variousmarginalized communities and indigenous people;and

• enhancing the capabilities of the poor and thevulnerable through greater access to self-help, localaccumulation, new skills and technologies.

The International Crops Research Institute for theSemi-Arid Tropics (ICRISAT) has 30 years’experience of service to the poor residing in the SATregions, which are one of the less-favored areas. Thebroad understanding and the village level data fromthese regions have equipped ICRISAT to betteraddress the problems and constraints prevalent inthese regions. Reflecting the ICRISAT’s philosophy of“Science with a Human Face”, its research strategyfocuses on the poor in the marginal areas of the SAT toempower, value and work on what matters to people;and to learn from them. This “people centred”approach led to the development of a people-orientedand impact-driven ICRISAT research strategy.

Understanding the underlying determinants ofpoverty in the SAT and the pathways to its alleviationis a major challenge. In this regard, ICRISATconducted a series of brainstorming workshopscommencing in 2000 as a component of its SATFutures initiative. It discussed the future of agriculturein the SAT and shared ideas about ICRISAT’s future

The last four decades have registered impressive gainsin food production, food security and rural povertyreduction in India. Increased intensification ofagriculture through intensive use of irrigation,fertilizers, and pesticides along with High YieldingVarieties (HYVs) in more-favored high-potentialzones was the major driving force for this success.However, many regions in less-favored rainfed areaslike much of the semi-arid tropics (SAT) have notbenefited from this process of agriculturaltransformation. Along with low productivity of rainfedagriculture and widespread poverty, the changingglobalized environment, scarcity of water, anddegradation of productive resources (land andbiodiversity) are threatening to further marginalizeagriculture and livelihoods in the Indian SAT.Stimulating productivity growth in the SATagriculture is the key for improving the well being of41% of the India’s poor now inhabiting the less-favored areas.

As opportunities for further growth in the more-favored regions are exhausted, food security andproductivity growth in agriculture in India areincreasingly depending on improved utilization ofnatural, human, financial, and material resources andproductivity growth in the less-favored regions. Theemerging evidence of higher impacts on poverty aswell as higher marginal productivity gains from publicinvestments in the less-favored regions suggests theneed to prioritize these hitherto overlooked areas interms of technology development and diffusion.

If future agricultural growth is to benefit the poorand contribute towards equitable economic growth, itis important to recognize the untapped potential ofSAT, and design suitable strategies and policies forstimulating sustainable productivity growth in theseregions. Adverse biophysical conditions and scarcity ofwater, which characterize much of SAT, and the widediversity and fragility of ecosystems in these regions,are likely to require approaches that differ from theGreen Revolution strategy. But simply introducingblanket policies targeted at increasing yields is notsufficient as past experience has shown.Understanding who gains from interventions and wholoses and why, is now known to be more important inaddressing persistent poverty in the SAT areas. Thiscalls for an interdisciplinary and crosscutting approachthat is long-term and sustained. The SustainableLivelihoods Approach (SLA) offers a good starting

6

Overcoming Poverty in Rural India

role in both Asia and Sub-Saharan Africa (SSA).Discussions were organized around the SLA in theprocess of identifying primary constraints andopportunities in SAT. This served as a basis for theseries of follow-up discussions with stakeholders andrelease of a book on “Future challenges andopportunities for agricultural R&D in the semi-aridtropics” published by ICRISAT in 2001. This wasadapted and expanded to capture the essentialdynamics of poverty for the design of its village levelstudies and measurement of research impacts anddevelopment of the ICRISAT’s vision and strategy to2010.

A poverty diagnostics and strategic assessment ofthe issues confronting the SAT is intended to furtherdeepen our understanding of the linkage between ruralpoverty and SAT, and the nature and dynamics ofpoverty. This would help design appropriateinterventions to help the poor achieve preferredlivelihood outcomes. Besides, this would alsocontribute significantly to IFAD’s understanding ofthe opportunities for the SAT agriculture to identifykey interventions and priority investments.

1.2 Importance of SAT and SATagriculture in India’s economyThe semi-arid tropics (SAT) cover parts of 55developing countries. There are many definitions ofthe SAT (see Appendix 1.1). In this report, the SAT isdefined using two parameters, ie, length of growingperiod (75–180 days) and mean monthly temperature(more than 18 oC). Thus defined, the SAT extends

Figure 1.1. Agro-ecological zones of India.

over a total geographical area of nearly 1.2 million sqkm, which constitutes 37.2% of the total geographicalarea of India (Figure 1.1 and Table 1.1). It is home tonearly 37% of India’s total population. In 1997–98,the SAT areas accounted for 46.2% of India’s total netcultivated area and 31.9% of gross irrigated area,58.7% of the area under coarse cereals, 59.7% of thearea under oilseeds, 52.6% of the area under pulsesand 60% of the area under commercial crops such as

Table 1.1. Relative importance of SAT vis-a-vis other climatic zones of India: 1997–98.

Share to All India (%)*

Importance/region Arid Humid Semi-arid temperate Semi-arid tropics

Geographical area 10.3 22.3 12.0 37.2Population 4.3 32.3 21.5 36.9Net cultivated area 11.4 21.1 19.0 46.2Gross cropped area 11.3 22.6 21.1 42.9Gross irrigated area 12.4 16.4 35.1 31.9Coarse cereals area 16.9 8.0 16.5 58.7Pulses area 12.9 17.4 17.1 52.6Oilseeds area 11.9 10.7 14.6 59.7Commercial crops area 15.5 8.2 16.3 60.0Fruits and vegetables area 2.3 49.9 19.2 28.6Production of coarse grains 10.0 10.3 19.2 60.5Production of pulses 7.7 16.0 24.8 51.5Production of oilseeds 10.0 10.1 13.5 62.8VOP agriculture 8.7 27.8 26.6 36.9

*Figures do not add to 100 because of unavailability of data from a few states.Source: ICRISAT database

7

Introduction and background

cotton and sugarcane. Similarly, it made a significantcontribution to the food grains basket of the country.For example, in 1997–98, it contributed 60.5% oftotal coarse grains production, 51.5% of totalproduction of pulses, and 62.8% of total oilseedsproduction in the country.

Several areas that are classified under the SAT inIndia are likely to be similar to what Scoones andThompson (1994) describe as the “Third Agriculture”,which is complex, diverse and risk prone, and yetsupports 1.9–2.2 billion people globally. The SouthAsian SAT, home to nearly 300 million people, covermainly southern interior areas of India and small partsof Pakistan and Sri Lanka (Kerr 2000). Poverty levelsare high and literacy rates are low. With productivitygreatly limited by low and variable rainfall, lack ofirrigation, and poor soils, the SAT agriculture contrastssharply with irrigated Green Revolution areas.Appendix 1.2 provides a list of predominantly the SATdistricts in India. The relative importance of the fourregions is detailed in Appendix 1.3 for two periods,1970–71 and 1997–98.

1.3 Important interventions inSAT agriculture and their impactsIn India, rainfed farming has been practised sincetimes immemorial. India has a long history, perhapsthe longest in the world, of government intervention indryland agriculture with a view to increase andstabilize crop yields. A few of those interventions arelisted below:• Establishment of dryland research stations in the

1930s;• Establishment of the Central Soil and Water

Conservation Research and Training Institute atDehradun in 1954 and its regional centers;

• Establishment of the Central Arid Zone ResearchInstitute (CAZRI) at Jodhpur in 1959;

• Launching of an All-India Coordinated ResearchProject for Dryland Agriculture in 1970;

• Launching of five Operational Research Projects in1975 and another four in 1984;

• Establishment of the Central Research Institute forDryland Agriculture (CRIDA) in Hyderabad (AP)in 1984;

• Development of 47 model watersheds in the 1980sand 1990s; and

• Launching of the National Watershed DevelopmentProgramme for Rainfed Areas in 1990s.In addition, several national level programs such as

the Drought-prone Area Programme (DPAP), DesertDevelopment Programme (DDP) and povertyalleviation and employment generation programs were

also launched by the Government of India with themain objective of development of dryland agricultureand reduction in poverty and unemployment. Theestablishment of ICRISAT at Patancheru in 1972significantly complemented the national efforts andinitiatives.

However, despite all these interventions, the SATagriculture did not register as much progress asagriculture in the irrigated areas elsewhere in thecountry and incidence of rural poverty remained high.The crop yields continued to be lower than in irrigatedareas and so also the farm incomes. Some of the majorchanges in the SAT agriculture, which could beattributed to the interventions made, are brieflysummed up below:

Increase in land productivity: Over the period, 1970–73to 1992–95, there had been several changes in the areaand productivity of major dryland crops. For example, theaverage yield of coarse cereals had increased by 65.4% andthat of oilseeds by 57% (Kanwar 1999). Most of thegrowth occurred in the 1970s in the irrigated areas and inthe 1980s in the rainfed areas. The total factorproductivity actually declined in both the areas in the1990s. The average value of crop and livestock outputswas Rs 21,985 ha-1 in the SAT during 1997–98 while itwas Rs 31,563 ha-1 in the non-SAT regions (Table 1.2).

Change in cropping pattern: Gulati and Kelley (1999)in their study found that there had been a distinct shiftin the cropping pattern in the SAT over the period,1968–70 to 1992–94, away from coarse grains in favorof wheat, paddy, and oilseeds. Using more recent data,we find that the share of oilseeds in the gross croppedarea of SAT had gone up from 11% in 1970 to 19% in1997–98. In contrast, the area under pulses declinedfrom 21 to 16%, and the area under coarse cerealsfrom 31 to 21%. With the expansion of irrigation,wheat has emerged as the favored crop.

Increase in area under irrigation: There has been amarked increase in the Gross Irrigated Area (GIA) inthe SAT from 12.5 million ha to 23 million ha over theperiod, 1968–70 to 1997–98. Much of the expansionin irrigation has occurred in the northern regions,facilitating shifts to high value crops such as wheat,paddy, sugarcane and rapeseed/mustard. Irrigation hasalso led to an increase in the intensity of cropping inthe SAT from 1.10 to 1.25. Yet, 70% of the grosscropped area remains rainfed.

1.4 Objectives of the study andresearch questionsThe main objective of the study is to providesignificant inputs for development of a Strategic

8

Overcoming Poverty in Rural India

Opportunities Report for India by IFAD. Inparticular, the study is designed to (1) establish alinkage between the SAT agriculture and ruralpoverty; (2) understand the critical issues of SATagriculture in India; (3) recommend strategies toincrease the contributions of SAT agriculture inreducing poverty; and (4) recommend prioritydevelopment interventions and research directions.The detailed terms of reference of the study are givenin Appendix 7. More specifically, the study seeks tofind answers to five questions: (1) Who are the ruralpoor? (2) Where are they concentrated in the SAT, orelsewhere? (3) Why are they poor? (4) What has beendone to reduce rural poverty? (5) What should bedone to further reduce rural poverty?

1.5 The sample and sources ofdataThe study is based on a review of the relevantliterature readily accessible to the study team and thefollowing sources of data:(a) 55th Round of the National Sample Survey (NSS)

conducted during the period 1 June 1999 to 31July 2000, covering 537 districts of India and asample of nearly 70,000 rural households;

(b) Population Census 2001;

(c) ICRISAT database for 1970 to1997 covering 483districts;

(d) CMIE (Centre for Monitoring Indian Economy)database; and

(e) Longitudinal Village Level Studies of ICRISAT inselected villages in Andhra Pradesh and Maha-rashtra.

1.6 Structure of reportThe study is structured to include the followingcomponents: Chapter 1 provides an overview of thestate of SAT agriculture in India and Chapter 2presents an analytical framework for the study basedon essential principles of sustainable livelihood.Chapter 3 identifies linkages between rural povertyand the SAT agriculture. Chapter 4 presents majorlivelihood strategies and current interventions in theSAT agriculture. Chapter 5 addresses the majorissues affecting the SAT agriculture in India,including biophysical, technical, economic andinstitutional issues. The report concludes in Chapter 6with a set of recommendations on strategies toincrease the contribution of SAT agriculture inreducing rural poverty in a sustainable manner. Thisalso identifies and highlights priority thematic andgeographical areas for possible interventions as wellas priority research directions for future.

Table. 1.2. Selected indicators of SAT and Non-SAT districts of India: 1997–98.

Indicators SAT Non-SAT All India

Population density (No/sq km of geog. area)1 300 393 324Literate rural female (%)1 39 38 39Urban population (%)1 30 22 28Average size of land holdings (ha) 1.9 1.4 1.6Proportion. of small land holders (%) 69 81 78Net area sown per capita 0.18 0.14 0.15Livestock units (units/ha cropped area) 1.7 2.2 2Livestock units per capita 0.33 0.29 0.31Gross value of agricultural produce (Rs/capita) 3919 4030 3989Gross value of agricultural produce (Rs ha-1) 21,985 31,563 27,209Tractor density (No./000 ha) 4.5 11.7 8.3Diesel and electric pump sets density (No./ ‘000 ha) 74.2 75.5 74.9Cropping intensity (%) 124.2 142.9 139.2Area under HYVs (%) 31.7 41.6 40.7Gross Irrigated area (%) 28.9 41.7 38.3Fertilizer consumption (kg ha-1 of NCA) 97 117 101Fertilizer consumption (kg ha-1 of GCA) 78 83 85Mean annual rainfall (mm) 965 1364 1212Market density (No/10,000 km2) 21.3 22.9 22.1Road density (km/sq. km) 0.5 0.5 0.75

1. Based on 2001 Census dataSource: ICRISAT databaseNCA: Net cropped areaGCA: Gross cropped area

9

2. Sustainable Livelihoods Approachand Analytical Framework

2.1 IntroductionThis chapter provides a conceptual and analyticalframework that underpins the analysis and synthesisof results that are presented in subsequent chapters ofthis report. This study uses the principles ofsustainable livelihood in developing its overarchingframework. First, we expound on the genesis, evolutionand current thinking about Sustainable LivelihoodsApproach (SLA). Adaptation of the principles of SLA isintended to guide the understanding and analyses ofpoverty and livelihood strategies in SAT. The SLA isadapted and expanded to capture the essentialdynamics of poverty in the process of identifyingprimary constraints and opportunities in SAT. Inparticular, the expanded framework added emphasison the aspects of markets, access, agricultural researchand technology, and power. It recognizes the multi-dimensional nature of poverty, encompassingeconomic, social, and governance perspectives. Thischapter also presents the basic concepts and measuresof poverty, including a specification of determinants ofincome poverty.

The context of vulnerability in the SAT agriculture isbased on a relatively risk-prone environment character-ized by low and erratic rainfall, and poor soils. In thisway, several groups of people who live within the SATareas are poorer and more vulnerable than people inbetter-endowed regions. Understanding how ruralhouseholds within the SAT manage risk, covariant riskor risk that affects everybody such as the risks imposedby drought, as well as idiosyncratic risks, which affectparticular household alone through coping, adaptationand diversification strategies is important. A key featureof the SAT livelihood strategies is the substitutionbetween different kinds of assets (human, political,natural etc.) and activities in response to the context ata particular point that allows them to deal effectivelywith risks imposed by a highly variable environment. Infact this goes a long way in explaining why many SATareas have demonstrated resilience in the face ofadverse environmental conditions that defy doomsdayprophesies and orthodox predictions about thefiniteness of resources.

2.2 Sustainable livelihoodsapproach and framework

The SLA grew out of a number of analytical streams,from diverse disciplines including economics,sociology, anthropology and the new institutionaleconomics. Amartya Sen’s theory of entitlements(Sen 1981) highlighted the importance of access toresources and challenged the notion that resourcedeprivation occurred as a result of production failurealone. Sen’s theory showed, for example, why a richfarmer with land, animals and other assets would beable to command more resources compared to alandless laborer for whom the only endowment ortradable asset is often labor. Sen later developed thiswork further with the idea of capabilities (Sen1990), which arose from the recognition thatincreased access to a resource can improve the welfareof the users only if they are capable of utilizing iteffectively. Swift (1989) broadened the definition ofassets owned by an individual to include a wide rangeof tangible and intangible stores of value or claims toassistance which can be mobilized in a crisis.

Davies (1996) distinguishes between ex-ante andex-post coping strategies. She likens ex-ante copingstrategies to insurance strategies because they areundertaken in anticipation of the shock. Ex-poststrategies are undertaken in response to the event orshock. In the long term, people may engage in moreadaptive strategies.

Robert Chambers’ theory of vulnerabilitycontinues to be a core influence in livelihoods thinking.Vulnerability occurs as a result of external risks, shocksand trends1. It is also because of defencelessness, whichstems from discrimination on the basis of class, age,gender, ethnicity, and disability. Vulnerability is not thesame as poverty, although the terms have been usedinterchangeably by some, and it is crucial to understandthe meaning because this helps us in understanding thecircumstances of marginalized groups.

Studies on gender have highlighted entrenchedinequalities in control over assets, genderdiscrimination in labor markets, and lack of voice in

1. Although the theory allows for a broad interpretation, the vulnerability context is usually interpreted in such a way that it underplays the importance ofliberalisation and conflict [Chronic Poverty Programme Website].

10

Overcoming Poverty in Rural India

the power struggles controlling resource allocation asthe main reason for women’s vulnerability. Genderresearch has also created awareness of thefeminization of poverty (Jackson 1995; Kabeer 1995;Lockwood 1995) and the often vulnerable situation offemale-headed or female-maintained households(Buvinic and Gupta 1994).

The New Institutional Economics (NIE) highlightsthe importance of formal and informal institutions aswell as transaction costs involved in the access toassets. Institutions are the rules of the game and areseparate from organizations and individuals who areplayers in the game.

2.3 A new sustainablelivelihoods framework for SATagricultureWe propose some additions (Figure 2.1) to the classiclivelihoods framework based on emerging concerns,especially those that relate to the SAT agriculture.Three areas need to be mainstreamed in livelihoodsanalyses, namely, markets, agricultural research andtechnology, and power.

Markets: In its classical form, the SLA is weak onmarket analysis, although as Dorward et al (2003)point out, this is a strange omission considering thatone of the main influences was Sen’s theory ofentitlements. Inadequate attention to markets canlead to failure in identifying and acting on a wide rangeof market, institutional and technologicalopportunities and constraints (Dorward et al. 2003).A more explicit analysis of the interactions betweeninstitutions, technology and assets in livelihoodsanalysis is required because the poor are alreadyengaged in a range of markets. In fact, some of themain constraints to livelihoods may arise from missingmarkets or imperfect markets, although as we will seelater, some have argued that the latter should beviewed as vital institutions that fulfill an importanteconomic function. Markets are more likely to beimperfect in remote areas, which often lie within theSAT boundaries.

Understanding the functioning of markets for labor,primary commodities, agricultural inputs and cheapconsumer goods is therefore crucial. Interventionscould then be formulated to ensure that newarrangements for trade and exchange do not createanti-poor markets and institutions and that the pro-

Figure 2.1. A sustainable livelihoods framework for SAT agriculture.

Vulnerability Context

Covariant RiskTrendsSeasonality

Water as an entry pointHigher public investmentHigher flow of institutional creditMixed crops livestock & trees farmingInfrastructure developmentCrop & livestock insurance

Policies, Institutions and Processes

GovernmentPoliciesMarketsInstitutionsAgricultural ResearchTechnologyConflictPowerSocial relations

Idiosyncratic RiskIntra-household relations

AssetsHumanPhysicalSocialNaturalFinancialPolitical

Positive LivelihoodOutcomes

More incomeIncreased wellbeingReduced vulnerabilityMore options

LivelihoodStrategies

CopingAdaptationDiversification

Negative LivelihoodOutcomes

DebtDecreased wellbeingIncreased vulnerabilityDepletion of NR base

11

poor elements of existing markets/institutions areenhanced.

Agricultural research and technology: Linked to theunder-recognition of markets is an under-recognitionof the role of technology in promoting livelihoods2.Although the record of technology development forthe SAT is impressive, uptake has been poor. India inparticular has developed a vast and diverse bank of(station-) proven technologies for dryland farming andconservation over several decades. This has beenachieved through massive investment in research, withthe strategic objective of doing for the drylands whatthe Green Revolution did for the irrigated areas(CRIDA 1997, quoted in gatekeeper series). Asexplained throughout the remainder of this report,this is due to the various constraints faced byhouseholds in the SAT areas. An SLA can help us tounderstand how agricultural research and technologyintersect with livelihoods and why they are adoptedsuccessfully in some cases and why not in others.

We draw on the work of Adato and Meinzen-Dick(2002) here and incorporate agricultural research andtechnologies in the policy and the institutions. Bothresearch and technology are strongly linked to theasset base as well as the vulnerability context.Technology, for example, irrigation, pest managementand livestock vaccinations can reduce vulnerability andthe vulnerability context can influence the choice oftechnology. Negative impacts of research andtechnology are also possible, especially if farmersadopt (for whatever reason) cultivars that are notsuited to their conditions as was seen in the case ofcotton in Andhra Pradesh, India. Certain kinds ofassets may be required to adopt new technologieseffectively as seen in the case of Green Revolutiontechnologies which were more suited to larger parcelsof land (natural capital) or those with access to credit(financial capital) and infrastructure (physical capital).Human capital, in the form of knowledge and skills, isnecessary to make proper use of many kinds oftechnologies.

Power: Recent feedback from applying livelihoodsapproaches highlights the inadequacy of SLA inaddressing power, politics and conflict both withinlocal communities and between groups of people andthe elite or formal institutions (Farrington 2001). Infact corrupt or rent-seeking officials and politicians aswell as upper caste people and the local elite maypresent the main obstructions to the poor in improving

their livelihoods. This realization has led certainagencies such as the World Bank and the Departmentfor International Development (DFID) to start takinga greater interest in corruption and strengthening ofpoor people’s capacity and voice to demand theirentitlements, for example, to fair wages and services.This aspect of livelihoods attains great importance inthe SAT areas which are often remote and wherepeople are more prone to exploitation throughinterlocked markets and less access to democraticinstitutions. Empowering them to fight for theirentitlements and rights presents special challengeswhich may be beyond the mandate of organizationsconcerned mainly with technology development. Butwe include it in our framework because it needs to berecognized as an important prerequisite for ensuringsuccess in the dissemination and uptake of severalkinds of technologies and improved agriculturalpractices.

An important dimension of vulnerability in remoteSAT areas is lack of power, voice, and social networks,which can help the poor to access resources,institutions, technology and markets. We thereforestress the significance of social capital and add a sixthtype of asset, which is “political” capital. This has beensuggested before in the literature (Bauman 2001) andmakes sense in the SAT context because this is where(a lack of) power is most likely to have an impact onlivelihood outcomes.

This report addresses in particular seven issues:identification of the poor, their livelihood strategies,access, trade offs, trends, technology and markets. Infacilitating the analysis, the analytical frameworkshown in Figure 2.2 addresses the five key questions:(1) Who are the rural poor? (2) Where are theyconcentrated? (3) Why are they poor? (4) What hasbeen done to reduce poverty? and (5) What should bedone to further alleviate poverty?

2.4 Multi-dimensional view ofpovertyThere are many quantitative measures of poverty thatare commonly used. Underlying most of these is theview that poverty connotes a state of affairs in which aperson or a household, or a community does not haveaccess to basic (minimum) necessities of life such asadequate food, clothing, shelter, healthcare, andeducation. One perspective is that a person is

2. Although the United Nations Development Programme (UNDP) approach does identify technology as one of the drivers of change.

Sustainable livelihoods approach and analytical framework

12

Overcoming Poverty in Rural India

Fig

ure

2.2

. A c

on

cep

tual

fra

mew

ork

fo

r u

nd

erst

and

ing

th

e p

oo

r, th

e p

ove

rty

and

th

e p

ove

rty

red

uct

ion

.

CONTEXT OF POVERTY

Liv

elih

ood

Stra

tegi

es

Who

are

Poo

ran

d w

here

are

the

yco

ncen

trat

ed?

Who

are

Poo

r?

•A

re th

eyla

ndle

ss?

•A

re th

eySC

/ST

?•

Are

they

illite

rate

?•

Are

they

wom

enhe

aded

hhs

?

Whe

re a

re th

ey?

•In

whi

chzo

nes

are

they

conc

entr

ated

•A

re th

ey n

ear

mar

kets

•A

re th

ey in

rura

l are

as•

Are

they

inm

argi

nal a

reas

•A

re th

eyev

eryw

here

•L

ow in

com

e

•L

ow h

uman

skill

s

•In

secu

reliv

elih

ood

•So

cial

exc

lusi

on

•H

igh

vuln

erab

ility

Wha

t ar

e th

eir

char

acte

rist

ics?

Why

are

the

yPo

or?

•B

ioph

ysic

al

•Te

chno

logi

cal

•E

cono

mic

•In

stit

utio

nal

•So

cio-

cult

ural

•A

dmin

istr

atio

n

•Po

licy

Wha

t is

bein

gdo

ne a

nd w

hat

has

wor

ked?

•C

opin

g st

rate

gies

•A

dapt

atio

n•

Div

ersi

fica

tion

•C

rop

prod

ucti

onch

oice

s•

Mar

keti

ng•

Non

-far

mE

mpl

oym

ent

•O

n-fa

rm n

atur

alre

sour

ce in

vest

men

t

Ext

erna

lIn

terv

enti

ons

•N

RD

•H

RD

•C

rop/

Live

stoc

k•

Pove

rty

alle

viat

ion

prog

ram

mes

•In

fras

truc

ture

•A

dmin

istr

atio

n &

Man

agem

ent

•St

akeh

olde

rspa

rtic

ipat

ion

Targ

ets

•E

nhan

celiv

elih

ood

stra

tegy

for

rur

alho

useh

olds

•In

crea

se t

hepr

oduc

tivi

ty o

fst

aple

cro

ps

•E

nabl

eco

mm

unit

ypa

rtic

ipat

ion

•St

reng

then

copi

ngm

echa

nism

s

•R

ole

of p

olic

y/G

over

nmen

t

•Im

prov

e m

arke

tac

cess

POVERTY REDUCTION

Con

stra

ints

13

considered poor if his/her consumption expenditureor income is below some minimum level necessary tomeet the basic needs of life. This minimum level isusually called the “poverty line”.

Depending upon the extent of gap between his/heractual consumption expenditure, or income and thepoverty line, a person may be classified as marginallyor moderately poor and severely poor. Similarly, wemay also differentiate between chronic poverty andtransient poverty depending on whether the personhas been poor for a relatively long period of time oronly for a relatively short period of time. Fourquantitative measures of poverty, namely, (1)headcount ratio, (2) poverty gap, (3) squared povertygap, and (4) mean per capita expenditure, were used inthis study. They are elaborated in Appendixes 4 and 5.

Poverty goes beyond lack of income. It is multi-dimensional, encompassing economic, social, andgovernance perspectives. Economically, the poor are notonly deprived of income and resources, but also ofopportunities. Markets and jobs are often difficult toaccess, because of low capabilities and geographical andsocial exclusion. Limited education affects their abilityto get jobs and to access information that could improvethe quality of their lives. Poor health, because ofinadequate nutrition and health services, further limitstheir prospects for work and from realizing their mentaland physical potential. This fragile position isexacerbated by insecurity. Living in marginal conditionswith no resources to fall back on, shocks become hardor impossible to offset. The situation is made worse bythe structure of societies and institutions that tends toexclude the poor from participating in decision-makingover the direction of social and economic development.

Evidences from the literature suggest that the conceptsand indicators used to measure poverty do matter a lotin identifying the poor.

2.5 Complementary frameworkfor identifying determinants ofpovertyFor conceptualizing the relationship between povertyand factors affecting it, we complement thesustainable livelihoods framework with principlesfrom the neoclassical economic theory of productionthat help determine the optimum level of use of aresource and optimal allocation of resources, whichstates that a rational producer tries to maximize his/her private profit subject to the constraints ofunderlying production function and other economic,institutional, and social considerations. The theory isput into practice through a set of principles includingthe principles of diminishing marginal returns andequi-marginal returns.

As we have seen above, poverty is a multi-dimensional concept and accordingly there are severalconcepts and measures of poverty. Of all the conceptsof poverty, income or consumption poverty is thesimplest to understand and measure. Appendix 3provides an elaboration on the measurements ofincome poverty and its determinants. Appealing to theexpanded conceptual framework based on principlesof sustainable livelihood (Figure 2.1), the context ofpoverty, causal factors and suitable interventions areidentified to underpin the subsequent analysis ofpoverty and livelihood strategies in SAT.

Sustainable livelihoods approach and analytical framework

3. Levels and Determinants of RuralPoverty in the Semi-Arid Tropics of India

on the basis of the importance of crop and livestockproduction. They have not considered agro-ecologicalzones. Furthermore, irrigation level of 40% as the cut-off point in consideration of an area as dryland is toohigh. Initially, we started with 12 zones, whichcomprised four broad zones – Humid, Arid, Semi-AridTemperate and Semi-Arid Tropics (SAT). Each zonewas further divided into three classes based on threeirrigation levels, viz, less than 25% (low), 25 to 40%(medium) and above 40% (high). The analysis focuseson rural poverty, so that purely urban districts and afew districts where data were not available were notincluded. The number of zones were reduced from 12to 10. The arid and semi-arid temperate zones, withlow and medium irrigation comprising only of a fewdistricts, were clubbed together. Our analysis of tenagro-climatic zones considering both pictures providesa better understanding of the structure and determi-nants of poverty.

The analysis is based on ungrouped household dataof the 55th Round (1999–2000) of the NationalSample Surveys (NSSs) on Consumer Expenditureand Employment and Unemployment. The householddata is aggregated for ten agro-climatic zones asillustrated in Figure 3.1. The classification considerslength of growing period and levels of irrigation, andtakes into account the household level multiplierssupplied by the National Sample Surveys Organization(NSSO). In addition to the NSS database, district-wise data on agricultural statistics from the Ministry ofAgriculture and Cooperation and decennial populationcensus data, are also used. The study adapted theestimates of rural poverty lines developed by Deatonand Dreze (2002). These are based on the pricesimplicit in the NSS data. It is noted that the use ofthese poverty lines results in a lower estimate of ruralpoverty than the official estimates of India. Correla-tion coefficients between rural poverty and socio-economic factors are calculated using the district levelestimates of poverty derived from the NSS data. Thedistrict level estimates of poverty are likely to sufferfrom high standard errors because of the small samplesize at that level. However, their use as dependentvariable in regression analysis will not create a seriousproblem because even if the dependent variablecontains errors, the estimates will be unbiased,although variances will be high (Gujarati 1987).

3.1 IntroductionThe previous chapter highlights the importance of aholistic approach, which considers the multi-dimensional perspective of poverty. Poverty being amulti-dimensional concept is influenced by severalfactors. To understand the nature and dynamics ofpoverty and therefore find the ways for effectiveintervention, policymakers first need to understandhow it is distributed across geographic and socio-economic groups, what are its characteristics, andwhat factors determine it.

The literature is replete with the evidence thatrural poverty is strongly related to the structure andperformance of agriculture, especially in India. Timeseries analysis clearly shows positive associationbetween poverty decline and agricultural growth. Butthe relation between agricultural growth, rural povertyand agro-climatic conditions is not very clear. Thischapter establishes the association between ruralpoverty, agricultural development, and agro-climaticfactors using the National Sample Survey (NSS) dataof the 55th Round (1999–2000). The analysis focuseson three aspects: (1) Who are the poor? (2) Where dothey live? (3) Why are they poor? It also attempts toanswer the question: what has to be done to bring thepoor out of poverty. Starting with the concept ofincome poverty, the analysis is extended to otherindicators, correlates, and determinants of poverty.The concentration of the poor in various agro-climaticzones is discussed. Caste, occupation and gender arealso considered in the analysis.

3.2 Sources of data andmethodologyStudies on patterns of poverty in India generallyconsider state as the unit of analysis. Since, states arenot homogeneous in terms of agro-ecologicalcharacteristics, analysis at that level cannot captureand explain the effects of agro-climatic differences.What we need are relatively homogenous systems/zones with a distinction on the basis of irrigation levelsthat will provide a better understanding of theincidence, correlates and determinants of poverty. Fanet al. (2000) have analyzed the problem by classifyingdry areas (irrigation ratio less than 40%) into 13 zones

14

15

Levels and determinants of rural poverty in the semi-arid tropics of India

Since poverty is a multi-dimensional concept, thestudy extends the analysis to examine other aspects ofdeprivation where quantitative measures (such aseducation and health) are available. Using theindicators of standard of living, knowledge and health,an index of human development (HDI) is constructedadopting the methodology of United NationsDevelopment Programme (UNDP). Appendix 5provides a detailed description of the methodologyused for computing the measures of poverty.

3.3 Where are the poorconcentrated?The head count measure of poverty shows that 21.3%of the India’s rural population is poor in 1999–2000with levels of poverty incidence varying across zones(Figures 3.2 and 3.3). The estimates shown areactually lower than the official or the Deaton’sestimate as we have used Deaton’s poverty lineswithout correcting for overestimation as he did. Thepoor are concentrated in the humid and SAT zones.The incidence in each of these zones is about 24.0%

and each accounts for 40.5% of the country’s ruralpoor. Thus, the humid and SAT regions account for81% of the rural poor, though their share in totalpopulation is 72% (as illustrated in Figure 3.4). Thesemi-arid temperate zone at 14.6% and the arid zoneat 12.6% have a lower incidence than the other twozones. The semi-arid temperate zone accounts for14.5% of the rural poor in India, while its share in ruralpopulation is 21.1% (Table 3.1)*. The arid zoneaccounts for 2.2% of the rural population and 1.2% ofthe rural poor. The remaining 3.2% of the poor are in

Figure 3.1. Agro-ecological zones and levels of irrigation, India.

Figure 3.2. Incidence of head count povertyacross zones.

* All the tables referred in this chapter are given in Appendix 3.

16

Overcoming Poverty in Rural India

the unclassified areas. It is important to note that areaswith low irrigation have the highest incidence in all thezones. The reason for low incidence of poverty in thesemi-arid temperate zone is that this region, especiallymajor parts of Punjab and Haryana, have benefitedmuch from the Green Revolution. Less populationpressure, more non-farm activities and lessemployment in the agriculture sector in the arid regionappears to explain the relatively low incidence ofpoverty in arid region.

Variations in the incidence of poverty within thebroad zones reveal that areas of medium level ofirrigation have lower incidence of poverty than areas ofhigh irrigation in humid and tropical zones. In theother two zones, poverty is significantly lower in theareas of high irrigation as compared to the areas ofmedium irrigation. Further, the arid zone hasexceptionally low incidence of poverty even in the lessirrigated areas. The low poverty in temperate zone isbecause of its high average irrigation ratio. Forinstance, the irrigation level in the semi-arid temperatezone is 62.1%, as against 31% and 28.4% in the humidand SAT zones, respectively. Even in the high irrigationcategory, the irrigation ratio in the semi-arid temperatezone is significantly higher than in the other two zones.Further, there is an inverse relation between irrigationlevel and rainfall in the humid and SAT zones as aresult of which the differences across irrigated levelsare narrowed down. Because of the high rainfall, areasof moderate irrigation are showing better performancethan areas of high irrigation.

In the semi-arid temperate zone, rainfall is lower inthe less irrigated areas and difference in irrigation level

is very high. As a result, difference in poverty incidenceis also quite high. The high-irrigation arid zone hasexceptionally low poverty because of its very highirrigation ratio at 79.3%. The areas of low and mediumirrigation in the arid zone have higher irrigation ratiothan their counterparts in the humid and SAT.

The other measures of poverty, viz, poverty gap,squared poverty gap, Sen’s Index and monthly percapita consumer expenditure yielded almost the sameranking of the regions as the head count ratio. Thesemeasures further confirm that the difference betweenthe humid and SAT zones is not significant (Table 3.3).

Because of higher share in rural population andhigher incidence of rural poverty, the SAT and humidzones have high concentration of the poor. Out of147.5 million rural poor in the country in 1999–2000,119.9 million are concentrated in the humid and SATzones and another 16.8 million are concentrated in thesemi-arid temperate zone with high irrigation.

3.4 Who are the rural poor?The aim of this section is to examine who are the ruralpoor among various classes across social groups,occupational groups and gender groups.

3.4.1 Rural poverty and social groupsIt is generally believed that poverty is higher amongthe Scheduled Tribes (ST) and the Scheduled Castes(SC) as a result of which the concentration of thesegroups in certain agro-ecological zones results in ahigher incidence of poverty. The incidence of poverty isfound to be the highest among the ST at 39.6%followed by the SC at 28.56% in contrast to 16.4%among general population (Table 3.6). The ST aremainly concentrated in the humid and SAT zones which

Figure 3.4. Distribution of poor across zones.Figure 3.3. Poverty levels in India by climaticzone.

17

partly is the reason for high incidence of poverty in thesetwo zones (Figure 3.5). These two zones togetheraccount for 93.8% of the ST whereas they contain only72.0% of the rural population. Even within these twozones, the distribution of the ST population is quiteuneven. The less irrigated areas in the humid and SATzones have a high concentration of this social group. Inthese areas, the ST account for 61.9% of the totalpopulation while the share of these areas in the totalpopulation is about 30%. Thus, the concentration of theST and the high incidence of poverty in the humid andSAT zones, especially in the less irrigated areas requirespecial attention in the policymaking. The SC areevenly spread across all the zones with a slightly highconcentration in SAT. It is observed that the ST (whoare mostly farmers) are concentrated in less irrigatedareas, whereas the SC (who are mostly laborers) areconcentrated in high-irrigated areas. The correlationcoefficient between poverty incidence and the STpopulation is found to be high at 0.47 for rural India(Table 3.25). The correlation is significant especially inmedium and low irrigated areas of the humid region andless irrigated areas of SAT. As these are theconcentrations of the ST, the results clearly indicate apositive association between poverty and theproportion of ST population. In the case of SC, thecorrelations are positive and statistically significant inhigh irrigated areas of all the four zones. From the aboveobservations, we find that there is a higher incidence ofpoverty among ST and SC communities than in thegeneral population. Moreover, the ST and SCcommunities are mainly concentrated in low irrigatedzones of the humid and SAT.

3.4.2 Rural poverty and occupationalgroupsIncidence of rural poverty is calculated separately forfive occupational groups, viz, agricultural laborers,non-agricultural manual laborers, other non-

agricultural hired workers (non-manual), self-employed in agriculture, and self-employed in non-agriculture. Among these five groups, poverty is thehighest among agricultural laborers at 35.5% and thelowest among the cultivators at 13.9% (Table 3.7). Thelaborers engaged in rural non-agricultural activitieshave significantly lower poverty than the agriculturallaborers. The people engaged in self-employedactivities in non-agricultural sector also have a lowincidence, but they stand next to cultivators. Thus, theagricultural laborers in rural areas face the highestincidence of poverty. It is significant to note that theincidence of poverty among agricultural laborers ishigh uniformly in all the zones and it is distinctlyhigher in the less irrigated areas. The correlationcoefficients between poverty and the proportion ofthe agricultural labor population are positive andstatistically significant in all the zones except the high-irrigated areas in the SAT zone (Table 3.25). Similarly,the correlations between poverty incidence and theproportion of non-agricultural laborer populations arenegative and statistically significant indicating thatdevelopment of rural non-farm activities is animportant policy intervention to reduce rural poverty.

Cultivators as a class are highly heterogeneous asthe size of operated land varies significantly acrosshouseholds. Incidence of poverty declines as farm sizeincreases in the semi-arid temperate and SAT zones(Table 3.8). In the humid and arid zones, povertyincidence is higher among small farmers as comparedto marginal farmers.

3.4.3 Rural poverty and gendergroupsAnother socio-economic group that is likely to bevulnerable to poverty is female-headed households. Itis observed that in areas of low irrigation in the humidzone, incidence of poverty among female-headedhouseholds is 20.1% while the incidence among male-headed households is 26.3%. Similar result is observedin the areas of low irrigation in the arid and SAT also.On the other hand, in the areas of high irrigation in thesemi-arid temperate and arid zones, incidence ofpoverty is significantly higher among female-headedhouseholds as compared to male-headed households.While the incidence is 21.0% among female-headedhouseholds, it is only 13.0% among male-headedhouseholds in the semi-arid temperate zone with highirrigation. These results show that at the aggregatelevel, for all the zones taken together, there is nodifference between the male- and female-headed

Figure 3.5. Distribution of SC, ST and totalpopulation.

Levels and determinants of rural poverty in the semi-arid tropics of India

18

Overcoming Poverty in Rural India

households in the incidence of poverty. But in the twohigh poverty zones, incidence of poverty is loweramong female-headed households as compared tomale-headed households.

3.5 Why are they poor?The previous sections of this chapter reveal that ruralpoor are mainly concentrated in the humid and SATzones and they are mainly agricultural laborers, smallfarmers and ST. This section deals with the question:why are they poor? Since most of the Indianpopulation is dependent on agriculture sector,agricultural development promises a strong positiveinfluence on rural poverty. Crop output, wages, cerealprices and other factors relating to agriculture alongwith assessment of the impact of anti-povertyprograms, and human development index (HDI) areimportant to account for while studying rural poverty.

3.5.1 Agricultural development andrural povertyCrop output is computed for the major crops for theyear 1997–98 (Table 3.18). The value of livestockoutput is also computed for the same year. Cropoutput per hectare is found to be Rs 20,350 at thenational level. The low poverty semi-arid temperatezone has the highest value of output at Rs 28,284 ha-1

and the SAT zone has the lowest value of output atRs 16,417 ha-1. The humid zone has significantlyhigher productivity of land than the SAT zone with perhectare output of Rs 23,999. It is found that areaswith high irrigation have distinctly high value of cropoutput in all the zones. The medium and low irrigationparts in the semi-arid temperate, SAT and arid zoneshave significantly low value of output as compared totheir counterparts in the humid zone. The resultsindicate that medium and low irrigation areas in thesemi-arid temperate, SAT and arid zones deserveattention for accelerating agricultural development.

When the value of crop output per capita isconsidered, the ranking of the zones is completelyaltered. The high poverty in the humid zone despiteagricultural development is due to high pressure ofpopulation on cultivated land. While the density ofpopulation per hectare of cultivated land is four in thetropical zone, it is more than seven in the humid andtemperate zones. Thus, poverty is same in both humidand temperate zones, despite significant difference inthe level of agricultural development. While perhectare output is at a moderate level, per capitaoutput is the lowest in the humid zone at about

Rs 3100, as against Rs 4100 in the temperate andtropical zones. The correlation coefficient betweenpoverty incidence and agricultural output per hectarehas negative sign as expected and is statisticallysignificant in all the zones (Table 3.25). Livestockoutput per capita also has a strong negative associationwith poverty. But the correlation between poverty andper capita crop output is not very strong. Fertilizer useper hectare is highest in the semi-arid temperate zonefollowed by the SAT and humid zones. It is found thatin highly irrigated areas in all regions, especially in thehumid and SAT zones, the use of fertilizer is very high.The SAT zone has a very high level of fertilizer use inrelation to its crop output per hectare.

The correlation coefficients of poverty indices withcrop output per hectare have expected negative signsand are statistically significant in all the zones and fordifferent irrigation levels within a zone (Table 3.25).Crop output per capita has negative association withpoverty indices in all the zones, but the correlationcoefficients are significant only in a few agro-ecologicalzones. For instance, in the humid zone all thecorrelations are statistically insignificant. Livestockoutput per capita also has the expected negative sign inall the zones, indicating that livestock output is animportant factor influencing poverty incidence. Thecorrelations are as strong as in the case of crop outputper hectare. Market density has a negative correlationwith poverty indices in humid and semi-aridtemperate zones. The other two zones do not indicateany significant association between poverty anddensity of markets. The density of roads has theweakest association with the incidence of poverty.Combining all the infrastructure components, anindex of infrastructure is developed and it is found tohave a significant negative association with povertyonly in a few zones.

3.5.2 Agricultural wages, cereal pricesand rural povertyAgricultural wages and cereal prices exert stronginfluence on the incidence of rural poverty. The priceof cereals is the highest in the humid zone and withinthe zone the low irrigated areas have a distinctly higherprice (Table 3.11). The high and medium irrigationparts of the SAT zone also exhibit high price of cereals.Thus, cereal price is higher in the entire humid zoneand parts of the SAT zone. Though agricultural wagerates are high in the humid zone, cereal price is also thehighest in this zone. The SAT zone has low wage rateand high cereal price. Thus, high cereal pricecombined with low agricultural wage rates are

19

contributing to high poverty in the humid and SATzones. The price of cereals is relatively low in the semi-arid temperate zone, but agricultural wages are thehighest in this zone. Thus, the low incidence ofpoverty in the semi-arid temperate zone appears tohave been contributed by high agricultural wages(Green Revolution zone, high surplus, low prices, andhigher rural employment) and low prices of cereals(Table 3.11). Most of the correlations with cereal priceare not significant and have no expected sign (Table3.25). On the other hand, the correlations withagricultural wages have the expected negative sign andmost of them are also statistically significant.

3.5.3 Poverty and food compositionCalorie intake levels are very low in the SAT zonesfollowed by the humid zones (Table 3.10). The shareof cereals in the total calories intake is the highest inthe humid zone followed by the SAT zone.Conversely, the proportion of calories obtained fromnon-cereals is relatively high in the arid and semi-aridtemperate zones. In a relative sense, pulseconsumption is high in the SAT and semi-aridtemperate zones, while milk consumption is higher inarid and semi-arid temperate zones. Humid and semi-arid temperate zones report a relatively highconsumption of vegetables. It is significant to note thatnon-vegetarian sources such as meat, fish and eggprovide only 0.8% of the total calories. Theconsumption of these items is relatively high in thehumid zone as compared to all other zones. Monthlyper capita food expenditure is the highest in arid zonefollowed by the humid zone. The proportion ofmonthly per capita food expenditure to the totalmonthly per capita expenditure exceeds 60% in thesetwo zones (Table 3.11). Interestingly, the unit price ofcereals is also relatively high in these two zones.Equality in consumer expenditure as measured by theratio of the expenditure of the bottom quintile to thatof the top quintile is better in arid zone whencompared to the other three zones.

3.5.4 Poverty and work participationratesWork participation rates have been computed forchildren and adults separately. Child participationrates are computed for the 6–14 age group (Table3.13). While work participation rates vary significantlyacross the zones, their association with the incidenceof poverty is weak. The SAT zone with the highestincidence of poverty has the highest work participation

rate and the arid zone with the lowest incidence alsohas high participation rate. On the other hand, thehumid zone has low participation rate despite its highincidence of poverty. Female participation rates varysignificantly across the zones (Table 3.14). The SATzone has the highest female participation rate at 61.1%and the humid zone has moderate female participationrate at 41.9%. Thus, one of the high poverty zones hashigh participation rate and one of the low povertyzones also has high participation rate. Similar is thecase with child participation rates. It is striking to findthat the arid zone has the highest incidence of childlabor at 17.6% and girls are contributing for this highincidence. While the participation rate for boys is14.3%, the corresponding rate for girls is 21.4%. TheSAT zone also has a very high incidence of child laborat 12.9%. On the other hand, the semi-arid temperatezone has the lowest incidence at 5.7%. Thesedifferences in the child work participation rates areclearly reflected in their participation in education. Theproportion of children attending educationalinstitutions is the lowest in the arid zone followed bythe SAT zone (Table 3.15). For instance, only 46.2% ofthe girls in the 6–14 age group are attending schoolwhile 61.6% are attending school in the semi-aridtemperate zone. These differences are also reflected inthe primary and secondary education completion rates(Table 3.16). Only 37.2% of girls in the 12–14 agegroup completed primary education in the arid zone ascompared to 55.3% in the semi-arid temperate zone.The completion rates of secondary education areabysmally low in all the zones and the differencesacross zones are narrowed down at this level.Unemployment rates computed on the basis of dailystatus in a week are the highest in the humid zonefollowed by the SAT zone (Table 3.14). Semi-aridtemperate zone has the lowest incidence of only 4.3%.Thus, unemployment is higher in high poverty areas ascompared to the other areas. The dependency ratiosare highest in the semi-arid temperate zone and lowestin the SAT zone (Table 3.17). Humid and arid zoneshave intermediate position with respect to thedependency ratios.

3.5.5 Poverty and Public DistributionSystem (PDS)

The Public Distribution System (PDS) is expected tosupply food grains and other commodities to thepoorer consumers at a subsidized price. Our analysisrevealed that only 12.4% of rice and wheat, 34.8% ofsugar and 76.5% of kerosene purchased by the ruralpeople are drawn from the PDS (Table 3.19).

Levels and determinants of rural poverty in the semi-arid tropics of India

20

Overcoming Poverty in Rural India

Purchases from the PDS are the lowest in the semi-arid temperate zone because the differences betweenPDS and open market prices are the lowest in thiszone in the case of rice, wheat and sugar (Table 3.20).This zone also has low incidence of poverty. In theremaining three zones, the share of PDS is about 15%.Though poverty is low in the arid zone, the demand forcommodities supplied through PDS is high because ofthe low production of rice and wheat. To get theproper indication of coverage, the participation of thepoor and the leakage through supply to the non-poorhave to be examined. The gap between the PDS priceand the open market price determines the demand forthe PDS commodities. Accessibility to the PDS isreflected in the purchases of the poor and the non-poor. The low purchase in the semi-arid temperatezone is due to lack of demand. This is partly because ofsmall gap between the PDS price and the open marketprice. The PDS price is lower than the market price byonly 30% (Table 3.20). Nearly two-thirds of the ruralpoor in the semi-arid temperate zone are notpurchasing from the PDS. There is also not muchleakage in the supply of grains through PDS in thiszone as less than 5% of the non-poor are getting cerealsunder the PDS. On the other hand, about one-half ofthe rural poor in the humid zone are not buying riceand wheat from the PDS. In contrast, two-thirds ofthe rural poor in the SAT and arid zones are purchasingrice and wheat from the PDS (Table 3.21). Except inthe semi-arid temperate zone, considerable sections ofnon-poor were also found to be purchasing rice andwheat from the PDS.

3.5.7 Human Development Index (HDI)

Since poverty is multidimensional in nature, one has togo beyond the measure of poverty estimated based onthe household consumption expenditures. Indices areconstructed for standard of living, knowledge, andhealth for each zone. An index of human developmentis computed as a simple average of the three indices. Inthe foregoing discussion, we observed that the humidand SAT zones have relatively high levels of povertywhen compared to the other two zones. The sameresults are reflected in the standard of living indices.Arid and semi-arid temperate zones have markedlyhigh levels of standard of living, but the resultspertaining to knowledge and health indices do notreflect the same trends (Table 3.24). Knowledge indexseems to be correlated more with the level of irrigationthan with the agro-climatic zones. But, in general,knowledge indices are higher for humid and semi-aridtemperate zones and are lower for arid zones, with theSAT zones occupying a middle position. Among thehumid sub-zones, both the high and low irrigated areashave much higher knowledge indices than the mediumirrigated zones. The same trend was also noted in caseof the SAT sub-zones. These results suggest thatknowledge indices are not necessarily correlated withthe standard of living indices. Arid zone, which had thelowest levels of poverty, also had very low levels ofknowledge indices. Health indices are showing muchmore diverse results. In the case of the semi-aridtemperate and arid zones, high-irrigated areas hadbetter health indices and low and medium irrigatedareas had relatively lower levels. But in the case of theSAT and humid zones, health indices had the highestvalues for low-irrigated areas. High-irrigated areas ofthese two zones occupied the middle positions withrespect to health indices and medium irrigated areasreported much lower values for health indices. Whilethe low-irrigated areas of SAT reported highest level ofpoverty, they showed highest measure of health index.There is no correspondence between standard ofliving, on one hand, and other social indicators such asknowledge and health on the other.

The HDI results show no clear trend or patternthat can be correlated with either the agroclimaticzone or the irrigation level. Arid high-irrigation zonehas the highest level of human development index andthe humid medium irrigation zone has the lowestmeasure. Medium irrigation zone areas seem to bereporting lowest levels of HDI in case of all agro-climatic zones. High irrigation areas in all the zones

3.5.6 Coverage of anti-povertyprogramsOnly 5.3% of the sample households were found tohave taken advantage of Integrated Rural Develop-ment Programmes sponsored by the governmentduring a period of five years (Table 3.22). Theparticipation in employment programs of thegovernment was even lower at 2.8% . These statisticsindicate that only 25% and 13% of the rural poor haveaccess to self-employment and wage-employmentprograms of the government, respectively. The ruralpoor from the humid zones had a higher coverageunder the anti-poverty programs than those in otherzones. The rural poor from the SAT and arid zoneshave relatively low utilization of the anti-povertyprograms.

21

and low irrigation areas of the humid and SAT zonesreported relatively high levels of HDI.

3.6 Key factors influencingpoverty and crop productivity

3.6.1 Determinants of PovertyThough incidence of poverty is relatively the same inthe humid and SAT zones, causes are likely to bedifferent and so also the policy interventions requiredfor reduction of poverty. In order to find the causes ofpoverty, an econometric model was estimated for eachzone using the cross-section data for the districts. Thisanalysis is confined to 410 districts because of datalimitations in the case of the other districts. Estimatesof poverty, unemployment rates, work participationrates, wages rates, cereal prices and literacy rate arederived from the 55th round NSS data on consumerexpenditure and employment and unemployment.Data for the remaining variables are drawn from thedistrict level database of the International CropsResearch Institute for the Semi-Arid Tropics(ICRISAT) which is based on the Ministry ofAgriculture statistics and 2001 Population Census.The value of crop output is estimated for 25 majorcrops and livestock output including the value of milk,meat, eggs and wool. The specification of the modelfor each zone is made on the basis of the expectedrelationships between poverty and hypothesizedcasual factors including agricultural and livestockoutput, wage rates, cereal prices, and infrastructureindex. The econometric specification used is in thelogarithmic functional form. The results of theestimated equations are shown in Table 3.26.

The model for all India contains all the importantdeterminants of poverty, viz, agricultural output perhectare, livestock output per hectare, agricultural andnonagricultural wage rates and unemployment rate. Aswe could see from the results of the regression analysispresented in Table 3.26, growth of crop output,reduction of unemployment and growth of wagescontribute independently to poverty reduction. Thereduction of poverty with agricultural growth in WestBengal, Rajasthan and Madhya Pradesh and reductionof poverty in Orissa without much agricultural growthbetween 1973–74 and 1993–94 has led to acontroversy regarding the association betweenagricultural growth and rural poverty (Mehta and Shah2003). Gaiha (1989) argues that growth alone is not

sufficient for poverty reduction. Employmentgeneration, effective PDS and better government arealso needed to alleviate poverty. The effect of wageincrease is much stronger than that of productivityincrease. An increase in the agricultural wage rate by10% reduces rural poverty by 1.19 percentage pointsand a similar increase in non-agricultural wage ratereduces rural poverty by only 0.59 percentage points.On the other hand, a 10% improvement in the cropproductivity reduces the incidence of poverty by 0.29percentage points only. Within agriculture, livestockproduction has a stronger effect on poverty than cropproduction. An increase in the per capita livestockoutput by 10% reduces rural poverty by 0.66percentage points. Though reduction in unemploy-ment reduces poverty, the effect is not very high. Theeffect of literacy is relatively low indicating that itsdirect effect on poverty reduction may be weak,although its indirect effect through occupationalmobility can be considerable.

The determinants of poverty differ significantlyacross zones. Livestock production turned out to beimportant only in low irrigated areas in the humidzone. These areas may be better suited to livestockproduction than others. However, if livestockproduction is further disaggregated into smallruminant and dairy production, the former may turnout to be important in SAT. Between per capita outputand per hectare output, the latter is found to beimportant only in the arid zone. This result should beinterpreted carefully. Output per capita rises onlywhen land productivity rises, given the constraints forarea expansion and the growth in rural population.Because of the favorable land-man ratio in the aridzone, growth in output per hectare is enough to reducepoverty.

The share of agricultural labor in total workers has astrong impact on poverty in all the zones. For any levelof agricultural development, a higher proportion ofagricultural labor leads to a higher incidence ofpoverty. But the semi-arid temperate and arid zoneswith low incidence of poverty are more sensitive tothe share of agricultural labor than other zones. Anyintervention in poverty alleviation should have thisgroup as the clearly identified target group. Increase inthe share of non-agricultural labor has a significanteffect on poverty in the semi-arid temperate and SATzones. The impact is stronger in the SAT with mediumirrigation than in the SAT with low irrigation. Itappears that agricultural development is a necessarycondition for the development of non-farm activity.

Levels and determinants of rural poverty in the semi-arid tropics of India

22

Overcoming Poverty in Rural India

It is observed that the SAT zone is agriculturallybackward and has moderate pressure on land, whichresults in relatively low value of crop output perhectare. Also, high cereal price combined with lowagricultural wage rates are contributing to high povertyin the SAT zones. Calorie intake levels are very lowand the share of cereals in the total calorie intake isvery high in the SAT zone. In a relative sense, pulseconsumption is high in the SAT and semi-aridtemperate zones, while milk consumption is higher inarid and semi-arid temperate zones.

SAT has a high incidence of work participationalong with a high incidence of female participation andchild labor. The proportion of children attendingeducational institutions is the lowest in the arid zonefollowed by the SAT zone. Unemployment ratescomputed on the basis of daily status in a week are thehighest in the humid zone followed by the SAT zone.The utilization of the anti-poverty programs by therural poor is less in the SAT and arid zones than in theother zones.

The results of regression show that while poverty iscaused by several reasons, low agricultural wages, highprice of cereals, high landlessness and highunemployment stand out as important. The low-irrigated areas in the SAT and arid zones may besuitable for small ruminant production than dairying.The SAT zone requires development of markets also.It should be noted that the semi-arid temperate zonehas excessive development of markets whereas theSAT zone is lagging in market infrastructure.

3.8 Synthesis of results andpolicy implicationsThe analysis throws light on three importantquestions: (1) Who are the poor and where are theyconcentrated? (2) Why are they poor? (3) Whatstrategies are needed to lift them from poverty? Ouranalyses revealed that the humid and SAT zones havethe highest number of rural population and togetheraccount for 72% of India’s rural population. Incidenceof poverty is high among the ST, the SC and theagricultural laborers. While the ST are concentrated inthe low-irrigated areas of the humid and SAT, the SCand agricultural laborers are evenly spread across allthe four zones. However, high-irrigated areas havehigher concentrations of the SC as well as agriculturallaborers. While poverty is caused by several reasons,low agricultural wages, high price of cereals, highlandlessness and high unemployment stand out asimportant. Their importance differs across zones. In

3.6.2 Determinants of CropProductivityImprovement in the value of crop output per hectaredepends on improvement in crop yields, increase inthe price of output, and shift in the crop patterntowards high value crops. Assuming that the croppingpattern is the result of constrained optimization by thefarmers, the scope for improvement in the value ofoutput per hectare is only through improvement ofyield. Yield levels are influenced by several factorsamong which irrigation ratio, fertilizer use, marketdensity and literacy rate are important. Exceptirrigation ratio and literacy rate, all other variables andthe dependent variable are introduced in logarithmicform. The equations are estimated for the four broadzones and all India.

The elasticity of output with respect to fertilizeruse is 0.12 at the national level and it is the highest inthe semi-arid temperate and SAT zones (Table 3.27).The elasticity for the semi-arid temperate zone isthrice and the elasticity in SAT is twice that of the allIndia average. The humid zone has very low fertilizerelasticity at 0.08. Irrigation ratio has significant impactin all the zones, but the effect is again negligible in thehumid zone. Market density has a positive impact onthe value of agricultural output only in the SAT zone.Density of markets is so high in the semi-aridtemperate zone that the coefficient turned out to benegative. The SAT zone requires policy interventionsthrough irrigation development, increase in thefertilizer use and development of markets. The humidzone has a greater potential for the development oflivestock sector than agriculture.

3.7 Major findings related toSATHaving analyzed the problems of rural poverty acrossthe four broad agro-ecological zones in a comparativeperspective, we focus on the structure and develop-ment of SAT. Poverty is more concentrated in this zoneas compared to others. While its share in ruralpopulation is 36%, its share in rural poor is 41.0%. Theregion has a high concentration of the weaker sections– 50% of the ST and 40% of the SC population of ruralIndia belong to this region. It is found that the SATregions have a high incidence of poverty (24.3%) andthe highest incidence of poverty was noted in the lowirrigation areas of SAT. The low irrigated areas in theSAT have a high concentration of ST among whom theincidence of poverty is found to be the highest.

23

the high poverty humid zone, agricultural wage ratesare high, but the cereal price is the highest among allthe zones, nullifying the advantage of high agriculturalwage. In the high poverty SAT zone, agricultural wagerates are low and cereal prices are high. The leakage inthe PDS is also high in these two high poverty zones.While 40 to 50% of the poor are unable to buy rice andwheat from the PDS, more than 35% of the non-poorpeople buy cereals from the PDS.

The two high poverty zones, ie, humid and SAT,differ in the level of agricultural development. Thehumid zone is agriculturally prosperous with a highvalue of crop output per hectare. But the pressure ofpopulation on land is very high. The SAT zone isagriculturally backward with a very low value of cropoutput per hectare. Because of the moderate pressureof population on land, value of crop output per capitais relatively low as compared to the other zones.Agriculture in the medium and low-irrigated areas inthe semi-arid temperate as well as the SAT zones hasto be strengthened. Improvement in agriculturalwages has a significant impact on rural poverty in theSAT and humid zones. For every 10% increase in rural

wage rate, poverty is likely to decline by 1.17 points inthe former and 0.77 percentage points in the latter. Areduction in the proportion of agricultural labor andincrease in the proportion of non-agricultural labor isalso expected to have a significant impact on povertyin these two zones. The low and medium-irrigatedSAT zone will gain in poverty reduction if per capitacrop output increases. The response of the low-irrigated humid areas to agricultural growth in terms ofpoverty reduction is not significant. But the responseto livestock development is quite high. The agro-ecological conditions of this zone are quite suitable forlivestock production. The low-irrigated areas in theSAT and arid zones may be suitable for small ruminantproduction than dairying. The semi-arid temperatezone requires rural diversification and plannedmigration to reduce the pressure of population onland. The elasticity of crop output for fertilizer use ishigher in SAT than in the other two zones. The SATzone requires development of markets also. It shouldbe noted that the semi-arid temperate zone hasexcessive development of markets whereas the SATzone is lagging in market infrastructure.

Levels and determinants of rural poverty in the semi-arid tropics of India

level findings. Six villages, two from Mahaboobnagardistrict in Andhra Pradesh and two each from Akolaand Solapur districts in Maharashtra, were intensivelystudied by the Social Science group of theInternational Crops Research Institute for the Semi-Arid Tropics (ICRISAT) during the period, 1975–76to 1984–85. The VLS were revived in 2002 and surveymethod was used to collect data from the originalsample households in these six villages. Four morewatershed villages in Andhra Pradesh were covered byVLS surveys in 2002. Analysis of income sources inthese ten villages helped identify important sources oflivelihood in the SAT villages. The estimates of averageannual net household incomes from 10 SAT villagesare presented in Table 4.1, which also containsinformation on the percentage of households with lessthan Rs 20,000 annual net household income (povertyline) along with the comparisons with the NationalSample Survey (NSS) estimates of the poverty in therelevant districts.

There is a considerable variation in the averageannual net household income between the villages,with Kanzara reporting highest income of Rs 60,687and Isthalapuram reporting lowest income ofRs 23,864. Same is the case with the percentage ofpoor households with annual income less than

4. Major Livelihood Strategies andCurrent Interventions in SAT Agriculture

4.1 IntroductionGiven the Sustainable Livelihoods Approach (SLA)and framework presented in Chapter 2 and the highincidence and severity of poverty in SAT and itsdeterminants as presented and analyzed in Chapter 3,we first examine in this chapter the major livelihoodstrategies of the SAT dwellers, and their strategies forcoping with natural or man-made shocks and threatsto their livelihoods. Thereafter, we critically evaluatethe effectiveness of current public interventions aimedat directly or indirectly reducing rural poverty andimproving the livelihoods of the SAT dwellers and thelessons of their experience that might be useful fordesigning future development and research strategiesfor SAT.

4.2 Livelihood portfolios andchanges over timeOur analysis in chapter 3 has established that theproportion of rural poor is higher in the Semi-AridTropics (SAT) region of India when compared withthe same in the arid, semi-arid temperate and humidregions. We have micro-level evidence from theVillage Level Studies (VLS) to substantiate the macro-

Table 4.1. Net household income levels and percentage of the poor in the SAT villages and relevantdistricts

Average percentage ofAverage annual % of households

the poor in the districtsName of household with annual income

S.No the village (District) income (Rs) <Rs 20,000 VLS estimate NSS estimate

1 Kanzara (Akola) 60687 6.0017.00 21.80

2 Kinkheda (Akola) 36606 28.003 Shirapur (Solapur) 51522 23.00

26.50 11.404 Kalman (Solapur) 43943 30.005 Aurepalle (Mahaboobnagar) 31561 33.006 Dokur (Mahaboobnagar) 36757 28.00 32.00 37.007 Sripuram (Mahaboobnagar) 36084 35.008 Karivemula (Kurnool) 26081 55.00 55.00 31.709 Nemmikkal (Nalgonda) 31796 45.00

50.50 17.8010 Isthlapuram (Nalgonda) 23864 56.00

Average (SAT) 37890 33.90 36.2 23.94

Source: ICRISAT VLS survey, 2001–2002

24

25

Major livelihood strategies and current interventions in SAT agriculture

from 40 to 27% over the 25-year period. The share oflabor income also dropped from 41 to 21%. There wasa marginal drop in the share of livestock income from11 to 10%. Non-farm income registered a growthvirtually in all the villages, registering, on an average, anincrease in its share from 4 to 23%. Caste occupationsand migration are contributing substantial amounts toincome in the Mahaboobnagar villages.

Due to water scarcity and land degradation, theproductivity and profitability of crops in the SATvillages are showing a decline. The Akola villages are anexception to this. The share of crops in the total nethousehold income has either remained constant or hasimproved. Income shares of labor and livestockdeclined in these villages, while that of non-farmincome increased. In the Mahaboobnagar villages,there was a drop in the share of crops in the total nethousehold income. This drop is quite sharp in the caseof Dokur village. The share of livestock in nethousehold income fell in Aurepalle, while it increasedsharply in Dokur. In both these villages, the share oflabor income decreased. In Solapur villages also, theshare of crops in the total net household incomedeclined. Same was the case with the labor income.The share of livestock in the net household incomeremained unchanged in Shirapur village, while itincreased substantially in Kalman village. In both thevillages, the share of non-farm income increasedsubstantially. Caste occupations and outmigration didnot contribute much to the net incomes of householdsin Maharashtra villages. The total net income of thehouseholds did increase in all the villages. The lowestincrease was noticed in Dokur, where it increased by1.92 times (or a 92% increase) over the income level in1975–78 converted to present value using consumer

Rs 20,000, which ranged between 6% in Kanzara and56% in Isthalapuram. The average annual nethousehold income in the 10 SAT villages was Rs37,890 and the average percentage of poor householdsworked out to be 33.90. In the last two columns, theaverage percentage of poor households in the fivestudy districts were compared with the correspondingNSS estimates. The percentage of poor was found tobe even higher in VLS sample (36.20%) than in NSSsample (23.94%). Part of the difference could be dueto the fact that whereas the VLS estimates are basedon income, the NSS estimates are based onconsumption expenditure. The major sources ofaverage annual net household income in the SATvillages are furnished in Table 4.2.

Farm income accounted for 53.28% of the averageannual net household income in the VLS villages. Butthe variation between villages was considerable. In theAkola villages with assured rainfall, farm incomecontributed more than 70% of the income, while inthe Dokur village, which was experiencing a persistentdrought for nearly a decade, it contributed a mere25%. On an average, non-farm income contributed27% of the total income. But its share was greater thanone-third in the villages located nearer to urbancenters, while its share was lower in the remotelylocated villages. Caste occupations and migration weremore important sources of income in villages ofAndhra Pradesh than in villages of Maharashtra.Migration and remittances contributed a significantshare of income in the villages where farm incomeswere lower.

Table 4.3 presents the changes in the structure ofincomes between 1975–78 (average) and 2001–2002in the six VLS villages. The share of crop income fell

Table 4.2. Sources of annual household income in SAT villages (in Rs).

Sources of income

Name of CasteS.No village Farm Non-farm occupation Migration Other sources Total

1 Kanzara 44737 (73.72) 8584 (14.15) 2364 (3.90) 942 (1.55) 4057 (6.69) 60686 (100)2 Kinkheda 26322 (71.91) 7418 (20.27) 656 (1.79) 562 (1.54) 1646 (4.50) 36606 (100)3 Shirapur 27314 (53.02) 18026 (34.99) 739 (1.44) 461 (0.90) 4980 (9.67) 51591 (100)4 Kalman 19832 (45.13) 18795 (42.77) 1168 (2.66) 100 (0.23) 4045 (9.21) 43942 (100)5 Aurepalle 12783 (41.00) 5983 (19.00) 9023 (29.00) 2396 (8.00) 1376 (4.00) 31561 (100)6 Dokur 9019 (25.00) 10156 (28.00) 4793 (13.00) 7491 (20.00) 5298 (14.00) 36757 (100)7 Sripuram 16663 (46.00) 11275 (31.00) 1181 (3.00) 2703 (7.00) 4262 (12.00) 36084 (100)8 Karivemula 14659 (56.00) 4535 (17.00) 1572 (6.00) 945 (4.00) 4370 (17.00) 26081 (100)9 Nemmikkal 16581 (52.00) 7414 (23.00) 3212 (10.00) 1621 (5.00) 2968 (9.00) 31796 (100)10 Isthalapuram 13981 (58.59) 9200 (38.55) 600 (2.51) 83 (0.35) 0 (0.00) 23864 (100)

Average 20189 (53.28) 10138 (26.76) 2531 (6.68) 1730 (4.57) 3300 (8.71) 37890 (100)

Figures in the parentheses represent percentages to the row totalSource: ICRISAT VLS survey, 2001-2002

26

Overcoming Poverty in Rural India

Tabl

e 4.

3. C

hang

es o

ver

tim

e in

the

sha

res

of d

iffe

rent

sou

rces

(%

) an

d le

vels

(R

s) o

f ho

useh

old

net

inco

me

in t

he V

LS

villa

ges.

Ave

rage

of

Aur

epal

leD

okur

Shir

apur

Kal

man

Kan

zara

Kin

khed

asi

x vi

llage

sSo

urce

s of

inco

me

1975

-78

2001

-02

1975

-78

2001

-02

1975

-78

2001

-02

1975

-78

2001

-02

1975

-78

2001

-02

1975

-78

2001

-02

1975

-78

2001

-02

Cro

ps29

.80

15.0

046

.10

3.00

33.7

022

.80

46.0

026

.40

43.9

052

.00

43.4

040

.00

40.4

826

.53

Lab

or32

.80

24.0

046

.30

14.0

042

.60

18.8

042

.10

18.0

038

.70

21.8

040

.80

30.0

040

.55

21.1

0Li

vest

ock

25.5

09.

002.

0018

.00

15.0

015

.70

0.80

6.50

9.00

5.70

13.1

02.

1010

.90

9.50

Non

-far

m11

.60

13.0

01.

2024

.00

0.20

31.8

04.

1037

.50

2.40

10.4

05.

3020

.10

4.13

22.8

0C

aste

occ

p.0.

0028

.00

0.00

6.00

0.20

0.30

0.00

2.20

0.00

1.90

0.00

1.80

0.03

6.40

Out

-mig

rati

on0.

008.

000.

0020

.00

0.00

0.90

0.00

0.20

0.00

1.60

0.00

1.50

0.00

5.37

Oth

ers

0.30

4.00

4.40

15.0

08.

509.

707.

009.

206.

006.

602.

604.

504.

808.

17

100

100

100

100

100

100

100

100

100

100

100

100

100

100

Tota

l23

6131

561

2967

3675

729

9551

390

1942

4394

338

5660

687

2522

3660

627

7443

491

(152

05)

(191

07)

(19

288)

(125

06)

(248

33)

(162

42)

(17

865)

Figu

res

in t

he p

aren

thes

es in

the

last

row

rep

rese

nt t

he e

quiv

alen

t cu

rren

t va

lues

of

base

yea

r in

com

es u

sing

who

lesa

le p

rice

inde

x at

the

all

Indi

a le

vel.

Sour

ce: I

CR

ISA

T V

LS s

urve

ys.

27

price index for agricultural laborers. Highest increasewas noted in the case of Kalman where thecorresponding increase was by 2.51 times (or a 151%increase over the base level).

Although income is an important indicator ofwelfare, consumption expenditure also reflects thestandard of living of people equally well. In case ofrural poor, expenditure on food grains accounts for asignificant part of consumption expenditure. Table 4.4contains data on consumption of food grains by thesample households in the VLS villages. On average, asample household consumed 799 kg of cereals and 61kg of pulses per year. In per capita terms, 153 kg ofcereals and 12 kg of pulses were consumed during2001–2002. The per capita food grain consumption of165 kg falls short of recommended consumption of180 kg per person per year. Besides the lower averagelevel of consumption, there is a considerableinequality in the consumption of food grains acrossvillages and income groups. These estimates ofconsumption lend support to the existence ofconsiderable undernutrition in the SAT areas.

Table 4.5 summarizes the data on consumptionexpenditure of the sample households and itsclassification into food grains, other food and non-foodcategories. The average annual consumptionexpenditure per capita was Rs 6510, which works outto only $0.39 per day per person. Considering the factthat World Bank considers a standard of 1$ per day perperson for drawing the poverty line, most of the

sample households in the SAT villages would obviouslyfall in the category of poor. Expenditures on foodgrains roughly accounted for one-fourth of the totalconsumption expenditure while another one-fourthwas spent on other food items. The remaining half ofthe expenditure was spent on non-food items. Thusthe analysis of consumption expenditure data showedthat there is considerable poverty as well asundernutrition among the sample households of SAT.

Table 4.4. Average annual consumption of foodgrains by the sample households in VLS villages,2001-2002 survey (kg).

Average Totalfamily food

S.No Village size Cereals Pulses grains

1 Aurepalle 4.2 747 56 8032 Dokur 5.4 742 50 7923 Sripuram 5.6 1039 45 10844 Kaivemula 5.6 1081 59 11405 Nemmikkal 5.1 1028 63 10916 Isthalapuram 4.5 852 37 8897 Kanzara 5.3 645 108 7538 Kinkheda 4.6 554 95 6499 Shirapur 5.5 630 41 67110 Kalman 5.7 668 62 730Average of VLS villages 5.2 799 61 860Per capita 153 12 165

Source: ICRISAT VLS Studies

Table 4.5. Average annual consumption expenditure of the sample households in the VLS villages (Rs).

Expenditures on

S.No Village Food grains Other food items Total food Non food Total

1 Aurepalle 7500 6948 14448 10288 247362 Dokur 8063 7042 15105 9429 245343 Sripuram 10798 6596 17394 15826 332034 Kaivemula 10578 5960 16538 11720 282585 Nemmikkal 10279 10789 21068 31025 520936 Isthalapuram 9373 6856 16229 18789 350187 Kanzara 6721 9618 16339 17331 336708 Kinkheda 5986 10129 16115 14639 307549 Shirapur 5650 11246 16896 19267 3616310 Kalman 6470 10417 16887 23179 40066

Average of VLS villages 8142 8560 16701 17149 33850

Average per capita 1566 1646 3212 3298 6510annual expenditure

Source: ICRISAT VLS surveys

Major livelihood strategies and current interventions in SAT agriculture

28

Overcoming Poverty in Rural India

Table 4.6. Distribution of families according to adoption of different coping strategies.

Percent of families adopting in

S No Coping Strategies Andhra Pradesh Gujarat Rajasthan

1. Use of food stocks/money savings 6.4 45.2 39.42. Purchase of low cost food items 61.3 73.3 31.03. Borrow cash/food from neighbors 63.5 63.7 51.44. Gather food from surrounding areas 2.0 2.3 2.85. Seek additional employment 2.7 10.5 12.26. Obtained Government assistance 10.7 8.3 5.17. Reduction in food consumption 23.3 47.3 35.68. Migration 39.1 10.0 8.19. Sell household or business assets to 12.7 3.8 9.4

good food or money

Source: Laxmiah and Vijayaraghavan, 2003

4.3 Coping strategies adoptedby farmers and agriculturallaborers

Besides low incomes, the SAT farmers and agriculturallaborers face too many uncertainties with respect totheir crop yields, prices and incomes. They facedroughts quite often and adopt many strategies tocope with the droughts and their consequences.Laxmaiah and Vijayaraghavan (2003) studied thestrategies adopted by households in coping withadverse situations, using the data collected from thirtyvillages each from the states of Andhra Pradesh,Gujarat and Rajasthan during recent drought years.The results are summarized in Table 4.6.

The most common coping strategies in AndhraPradesh were borrowing cash or food from neighbors,purchase of low cost food items, migration andreduced food consumption. In Gujarat, purchase oflow cost food items, borrowing cash/food fromneighbors, reduction in food consumption and use offood stocks/money savings were the dominantstrategies. In Rajasthan, the strategies were the sameas in Gujarat with a slight change in their rankings.This study concluded that the drought-affectedpeople mainly resort to borrowing, draw down ofstocks, reduced consumption, shifting to low cost fooditems and migration to survive the adverse fallout ofdroughts. The VLS of ICRISAT had a special moduleon the strategies adopted by the farmers to cope withthe adverse effects of droughts in one year andpersistent droughts for more than one year. Table 4.7presents information on the frequency and impacts ofdroughts and the strategies followed by the farmers tocope with the adverse effects of drought in the villagesof Andhra Pradesh.

There is approximately a 52% probability that adrought may occur in the SAT villages of AP. Theaverage shortfall in income because of a drought was44.2%. About two-thirds of the farmers adoptedalternative coping strategies to face the shortfall ofincome in a drought year. The most common copingstrategies when they face a single year drought areborrowing, shifting to non-farm labor work, reducedconsumption expenditure and migration.

In the SAT villages of Maharashtra, the droughtincidence is less frequent (25% probability) whencompared with the same in villages of Andhra Pradesh.The information on drought impacts and farmers’coping strategies are furnished in Table 4.8. Theaverage shortfall in income during a drought year wasabout 23% in villages of Maharashtra. Only 36% of thesample farmers reported that they were adoptingsome strategies to cope with the drought. In Solapurvillages, farmers participated as laborers in the workscarried out under the Employment GuaranteeScheme (EGS). Cutting down expenditures, changingthe cropping pattern and reducing the input use werethe other measures adopted by the farmers in Solapurvillages. Farmers in Akola villages participated in thelocal labor market during distress, but not in the EGSworks. Other coping measures were the same as inSolapur villages with slight difference in emphasis.When farmers face droughts for more than one year,they avail loans to dig more borewells, lease out theirlands, shift to non-farm occupations and migrate todistant places for work to meet their family needs.Other strategies of borrowing, reducing expenditures,and sale of animals and gold are also commonlyresorted to.

Diversification: Livelihood diversification is animportant adaptive strategy in the SAT areas for

29

Table 4.7. Impacts of droughts on farmers in AP villages and their response to cope with them.

Mahaboobnagar Kurnool NalgondaAverage

Sl No Parameter/Strategy Aurepalle Dokur Sripuram Karivemula Nemmikkal Isthlapuram (six villages)

1 No. of drought years 4.00 6.00 5.00 6.00 5.00 5.00 5.20in the last 10 years

2 Average shortfall in 44.50 55.70 40.00 55.00 35.00 35.00 44.20income because ofdrought

3 Percentage of farmers 74.70 88.30 72.50 70.00 47.10 47.10 66.20adopting copingstrategies

4 Percentage of farmers adoptingdifferent coping strategies

4(a) Shifted to non-farm 30.40 28.30 13.79 33.93 21.21 21.21 24.81labor work

4(b) Borrowed money 42.90 32.10 60.10 37.50 12.12 12.12 32.81

4(c) Sold draft animals/land 5.40 9.40 1.79 3.58 6.06 6.06 5.38

4(d) Shifted to dairy, toddy 8.90 9.40 3.45 1.78 3.03 3.03 4.93tapping, etc.

4(e) Migrated 3.60 11.30 10.33 12.50 21.21 21.21 13.36

4(f) Depended on 3.60 3.80 3.80 1.78 0.00 0.00 2.16old savings

4(g) Reduced consumption 5.20 5.70 6.90 8.93 36.37 36.37 16.55expenditure

Source: ICRISAT VLS survey 2001–2002

Table 4.8. Impacts of drought on farmers in Maharashtra villages and their responses to cope with them.

Solapur Akola

Sl No Parameter / Strategy Shirapur Kalman Kanzara Kinkheda Average

1 No.of drought years in the 3.00 3.00 2.00 2.00 2.50 last 10 years

2 Average shortfall in income 20.00 21.68 20.00 30.00 22.92because of drought

3 Percentage of farmers adopting 30.30 72.90 15.40 25.00 35.90coping strategies

4 Percentage of farmers adopting different coping strategies4(a) Cutting down expenditure 11.80 11.80 50.00 33.33 26.73

4(b) Participating in labor market 0.00 0.00 16.67 50.00 16.67

4(c) Changes in cropping pattern 9.80 9.80 16.67 0.00 9.07

4(d) Reducing the input use 3.90 3.90 16.67 16.67 10.28

4(e) Participation in employment 74.50 74.50 0.00 0.00 37.25guarantee scheme

Source: ICRISAT VLS survey 2001–2002.

Major livelihood strategies and current interventions in SAT agriculture

30

Overcoming Poverty in Rural India

raising incomes and reducing risk. Ellis (2000) defineslivelihood diversification as “the process by whichrural families construct a diverse portfolio of activitiesand social support capabilities in their struggle forsurvival and in order to improve their standard ofliving”. Diversification can mean two things althoughnot necessarily mutually exclusive; the diversificationof income sources including farm and non-farmsources and the diversification of rural livelihoodsaway from farming into the non-farm sector. Theconsensus so far is that the poorest and the richest arethe least diversified but for very different reasons: thepoor have fewer options and are constrained whereasthe rich can afford to specialize in a high returnactivity. Simplistic analyses have led to policyresponses, which seek to expand higher-return non-farm enterprises. But in-depth analyses have shownthat non-farm activities at the lower end may beerratic yielding low return because the poor lack theassets to engage in higher return activities (Start,Deshingkar and Farrington, Barrett et al., 2001;Reardon et al., 2000). In addition to this, several non-farm activities are closely dependent on agriculture.Therefore investing scarce resources in the creation ofnon-farm options to the detriment of improvingagricultural productivity may have negative conse-quences for the poor.

At the aggregate level, agro-ecological, socio-economic, technological and institutional factorsinfluence spatial patterns of diversification. These arebroadly classified into demand side and supply sidefactors. Demand side factors include per capitaincome, urbanization, tastes, and preferences, whichinfluence demand for diversified commodities. Majorsupply related factors influencing pattern ofdiversification are rainfall, technology (irrigation,fertilizer, High Yielding Variety (HYV) area,tractorization, improved breeds of livestock andpoultry), land area available for livestock production,scale of production and infrastructure availability suchas roads and markets.

As a matter of fact, agriculture in the SAT isdiversifying at a rapid rate. Between 1980–82 and1996–98, the share of cereals in total value of cropproduction declined from 51 to 40%, while the shareof oilseeds increased from 13 to 21%, and fruits andvegetables from 14 to 17%. During the same period,the share of crops in total value of agriculturalproduction declined from 84 to 80% but the share oflivestock products increased from 16 to 19%. Thelivestock sector has also been growing faster at 4.5%per annum compared to 3.3% for the crop sector. Inrecent years the share of high value commodities(HVCs) such as milk, meat, fruits and vegetables in

total agricultural production of the SAT is increasingmainly driven by demand side factors (ParthasarathyRao et al 2002) (see also box 4.1).

Their shares are higher than those for cereals,pulses and oilseeds. In 1980–82, these commoditiesaccounted for 28% of total value of agriculturalproduction (at 1980 prices) and their share has nowincreased to 33% in 1997–98. Small farmers are themajor beneficiaries of the growing demand for thesecrops. However, there is a need for vertical integrationof markets, by eliminating the intermediaries toincrease the producer’s share in the consumer rupee.

At this stage in the process of agriculturaldevelopment in SAT, diversification to high valuecrops such as dryland horticultural crops and tree-based farming systems seems to be a natural choice(Pareek 1999; Bhagmal and Roy 1999; Korwar 1999).Diversification to horticulture can reduce risks,improve profitability of cropping systems and ensuregreater nutritional security while the tree-basedfarming systems can reduce pressure on forest formeeting fuel and fodder requirements of the growingpopulation and in that sense, it improves sustainability.Generally, the production of horticultural crops isthinly spread throughout the SAT but some pockets ofconcentration do exist and offer potential for furtherdevelopment. However, any horticulture develop-ment program must be accompanied by the creationof markets, both in terms of physical infrastructureand consumer demand. Any attempt to increaseagricultural production and more particularlyhorticultural production without market developmentis bound to fail.

Some new thinking is leading to emergence of agro-forestry as a means of improving the productivity andprofitability of less productive and degraded lands.Agro-forestry means growing trees on farms toimprove livelihoods and protect the environment.Such agro-forestry systems are important to conservethe top soil and its moisture content required forsuccess of any cropping system. If successfullyadopted, this alternative may prove relatively moreattractive from the viewpoint of meeting fuel andfodder requirements of growing population, whichcontinues to remain dependent on natural forests andvegetation.

4.4 Current interventions in SATagriculture and their impactsAs we stated in Chapter 1, India has a long history,perhaps the longest in the world, of governmentintervention in dryland agriculture with a view to

31

Box 4.1. Livelihoods diversification – forward and backward linkages

Diversification was a strategy taken up by landless households and by small, medium and large farmers in thevillages of Aurepalle and Dokur in Andhra Pradesh. Those with large landholdings and productive assets werenot immune to the risks faced in agriculture. In fact, there was only a limited evidence of diversificationenabling households in Aurepalle and Dokur to accumulate wealth and assets in significant measures. Thosewho experienced an erosion of income and assets were then forced into the non-farm sector because therewere no opportunities for them in agriculture, except perhaps as very low-paid regular farm servants. Theinvestments made by others in irrigation and machinery, or the benefits accruing to people who received landunder distribution programmes offered a life-line to many households in the context of drought and cropfailure. Whilst some of the diversification strategies within and outside agriculture appear to have increasedincomes in real terms in the villages, diversification strategies are not themselves free of risk and, in theprevailing agro-economic climate, often offered little more than an opportunity to cope and mitigate risk or totread water and hold on to productive assets for the future. The findings beg an important question about theprocess of diversification in Aurepalle and Dokur and in the semi-arid tropics of India more generally. Whilstboth villages faced drought and a subsequent dearth of water for irrigation, it was not clear whether years ofdrought, and only average rainfall in intervening years, had brought about short-term or intermediate copingstrategies or a more meaningful and long-term change in the livelihood strategies of households. Given thatvery few households accumulated significant wealth through diversification, it may well be that, if futurerainfall is both plentiful and timely, then there will be a return to an overwhelming dependence on agricultureand agricultural labour, and a parallel decline in migrant labour and other non-farm activities. However, evenif there is a will to return to agriculture when improved rainfall conditions prevail, it also remains to be seenwhether households have, during the drought, disposed off too many of their agricultural assets to make aserious return to farming. Similarly, whilst in Aurepalle, population density declined between 1989 and 2001,population pressure in Dokur continues to increase. The population of both the villages increased between1975 and 2001. This also has implications for the future of agriculture since a continued rise in populationdiminishes the possibility of households gaining a livelihood from cultivation. The diversification process,coupled with uncertainty over availability of agricultural assets in the future, also raises important policyquestions. Above all, there remains a challenge for the structure in which government policy is made and stateinterventions are carried out. Whilst policy and interventions are implemented largely along sectoral lines,household livelihoods are highly diverse. How might the linkages between farm and non-farm livelihoods beexploited within existing policy channels to help generate new sources of livelihood? One appropriate strategyhere might be to encourage forward and backward linkages to agriculture by supporting enterprises that eitherenable better agricultural production (for example village repair services for agricultural machinery andimplements) or the process of adding value to agricultural production before it leaves the village (for examplemilling, food processing, packaging and transportation).

Source: Deb et al. 2002

increase and stabilize crop yields. A few of thoseinterventions were listed there.

The Government of India (GOI) have launchedseveral programs from time to time to restore,conserve and enhance the land and water resources inthe country; to alleviate rural poverty andunemployment; to improve the livelihoods of the ruralpoor; and to improve the basic infrastructure. Theseprograms could broadly be classified into the followingcategories:• Natural Resources Development and Management• Human Resources Development• Agricultural and Livestock Development

• Wage Paid Employment Generating Schemes• Self-Employment Schemes

A list of selected centrally-sponsored rural develop-ment programs currently underway in SAT, India, anda brief critique of some major programs focusing ontheir effectiveness/impact and the lessons derivedfrom their experience is presented in Appendix 6.

4.5 Concluding remarksThe sustainable livelihood approach and framework,augmented with the inclusion of issues relating to thepolitical economy of poverty, higher focus on markets

Major livelihood strategies and current interventions in SAT agriculture

32

Overcoming Poverty in Rural India

professionals, and redesign/refocus the programs suchthat their overall effectiveness and efficiency areenhanced. Better interventions are needed in theexisting production, marketing, research anddevelopment (R&D), institutional/organizational andadministration and management systems aimed atattaining food and nutrition security for everyone. Thiswould require, among other things, the diversificationof agriculture with introduction of high valuehorticultural crops and perennial tree crops (agro-forestry) and selective specialization keeping in viewthe prevailing agro-climatic conditions and availabilityof marketing infrastructure and facilities, adoption ofwatershed development approach for improving theproductivity of dryland agriculture through bettermanagement of land and water resources and usingappropriate technologies. Besides, enhancement offarmers’ access to well-functioning markets, andreorientation of existing R&D, institutions/organizations, and administration and managementsystems would also be needed.

and technology, presented in Chapter 2 seemappropriate to comprehend the problems of ruralpoverty and livelihood insecurity, to diagnose the rootcause(s) of the problems, and to identify potentialstrategies for their amelioration. The framework isholistic, multi-disciplinary and people-centered anddeserves to be applied in designing poverty reductionand livelihood improvement programs.

There have been many changes in the patterns andmeans of livelihood in the SAT region over the last 25years or so in response to changes in technology,infrastructure, markets, urbanization, and commercia-lization of agriculture. Diversification of sources ofincome and shifts in cropping patterns and food habitsare the ones that are occurring in most of the SATareas in India.

To reduce rural poverty and improve the livelihoodsecurity in SAT, there is a need for the governments tocritically review the performance and impacts of theexisting interventions, preferably with the assistanceof renowned rural development and management

33

5. Emerging Changes and Issuesin SAT Agriculture

rich and the poor, and industry and agriculture. Forexample, in many parts of Gujarat, conflicts betweenfarmers and urban dwellers over the re-allocation ofwater from irrigation reservoirs have in the recenttimes become more frequent, intense and occasionallyviolent (Ballabh and Singh 1997; Ballabh and Pradhan2000; Ballabh 2002).

Since time immemorial, droughts have been a baneof the SAT agriculture in India. They are considered asnature’s curse on human beings living in SAT, posing athreat to their survival and livelihood. On average, theSAT in India is prone to droughts once in every threeyears. The areas/states which are most vulnerable todroughts include western Rajasthan, easternRajasthan, Gujarat (Saurashtra, Kutch, and northGujarat), western Uttar Pradesh, Tamil Nadu,Kashmir, and Rayalseema and Telangana in AndhraPradesh. In these areas, droughts are expected tooccur every 2–3 years over a period of every five years(Singh and Ballabh 2002). Droughts have severalphysical, economic, social and environmental effectsand impacts, and cause immense misery to people andanimals living in the drought-affected areas. Loss ofagricultural production and incomes lead to severaldestabilizing effects on the entire economy of thecountry. Future research could focus on identifyinggenes (not only from current mandate crops, but alsofrom other species) that can improve water-useefficiency and confer drought tolerance. The researchagenda could also include crop and systems modeling,integrated watershed management, water policy, andinstitutional innovations in water resource trading,allocation, pricing and management.

5.2.2 Land degradation and poorquality of soilsLand degradation is a serious problem in India ingeneral and the SAT in particular. According to anestimate, nearly 107.43 million ha of land in India isaffected by various forms of degradation (GOI 1974).The soils have low water holding capacity and aredeficient in organic matter and several nutrients, andtherefore cannot support high crop yields on sustainedbasis. In view of this, one of the major challengesfacing the SAT agriculture in India today is to conserve

5.1 IntroductionEvidence from the literature indicates that sweepingchanges have occurred in the village economies of themore favored regions of Asia in the last few decades.The extent and broad-ranging implications of thesechanges justify an assessment of research anddevelopment (R&D) priorities in regions that havebeen bypassed. These developments suggest some keyissues and challenges confronting the SAT of India(Ryan and Spencer 2001). The SAT agriculture inIndia has not been as responsive to the emerging issuesand challenges as agriculture in irrigated areas of thecountry. This is mainly because it is beset with severalproblems. Some of the problems are intrinsic to thetype of agriculture extant in the SAT whereas othersare externally imposed. The problems could also becategorized according to their character such asbiophysical, technical, economic, institutional, socio-cultural, and so on. In this chapter, we first take stockof some of the emerging changes in agriculture andthen briefly describe the nature, extent and severity ofsome of the major issues confronting the SATagriculture in India and alternatives to resolve them.

5.2 Biophysical problems

5.2.1 Water scarcity and droughtsWater scarcity has been and still is the most criticalconstraint in the SAT agriculture. According to Ryanand Spencer (2001) “water will likely be the primaryconstraint throughout SAT in coming years (also)”.The problems of water scarcity and droughts in theSAT areas, as also in many areas of India, are to a largeextent, caused by the low and erratic rainfall, lack ofharvesting, storage and conservation of rain water,increased over-exploitation and pollution of bothsurface water and groundwater, lack of properallocation and efficient use of water, lack of well-defined property rights in water backed up by law, andthe shortcomings in the design and implementation ofdrought relief programs. Besides, its adverse effect oncrop yields, water scarcity also results in competitionfor and conflicts over water between states (eg,Karnataka and Tamil Nadu), towns and villages, the

34

Overcoming Poverty in Rural India

and enhance the inherent capacity of its land and othernatural resources to sustain agriculture. Any erosion ofthis inherent capacity will threaten India’s foodsecurity and sustainability of agriculture.

The SAT agriculture in India is prone to severe soilerosion and other forms of land degradation. Giventhe fragile nature of natural resources in SAT, there is aneed to develop technologies which do not create toomuch pressure on natural resources and which helpreplenish the resources used up in the process ofproduction. This will ensure that the stock of naturalresources is maintained intact over time. Research isneeded to understand the nature, extent, con-sequences, and trends in land degradation in SAT. Thisshould include the effects of soil loss and nutrientdepletion on productivity, water pollution, salinity, andloss of biodiversity.

5.2.3 High risk and uncertainty, andvulnerabilityThe SAT agriculture is characterized, among otherthings, by a high degree of risk and uncertainty causedby several weather-related aberrations such asabnormally low and erratic rainfall and its unevendistribution over time and space, droughts, hailstormsand so on. Given their income poverty, most of thefarmers do not have the capacity to bear such risks andhence become risk averse. This becomes a constraintin the adoption of new technologies by those farmers.Consequently, their farm production and incomes arelower than what is possible with the adoption of newdryland technologies and they become morevulnerable to such aberrations.

5.3 Technological issues

5.3.1 Low productivityThe SAT agriculture is characterized by low land andlabor productivity. It is lagging far behind the non-SATregions in India in terms of productivity of all majorcrops grown there (Table 4.1). For example, theaverage yield of wheat in 1997–98 was 1813 kg ha-1

compared to 2801 kg ha-1 for non-SAT region, cotton223 kg ha-1 compared to 305 in non-SAT regions.Although most dryland crops yields are marginallyhigher in the SAT compared to the non-SAT regionwith a few exceptions, the productivity levels are lowcompared to global averages for the dry land crops.The average crop productivity in value terms for majorcrops in the SAT is Rs 16,195 ha-1, while it is Rs 23,534ha-1 in the non-SAT region.

Besides, within the SAT region, there is a widevariation in crop yields. For example, the average yieldof sorghum varied from 370 kg ha-1 in farmers’ fields to1460 kg ha-1 in demonstration fields and 1060 kg ha-1

on farmers’ fields with assured input supply (Kanwar1999). This shows that there exists a lot of potentialfor increasing crop yields in SAT. In view of this, whatis required is greater investment in R&D to developregion-specific crop varieties keeping in mind the needfor bio-diversity conservation. In addition to steppingup the investments in R&D, it will also be necessary toensure that the research findings reach the targetgroup and that an enabling environment is created forfarmers to be able to adopt the latest research findingsso that the R&D impact is maximized.

Fan et al. (2000) have reviewed the performance ofIndia’s agriculture in irrigated and rainfed zones overthe period, 1970–1994. They found that the averagegrowth in total factor productivity in the irrigatedareas over this period of time as 0.92%, which wasslightly higher than in the rainfed zones (0.73%). Mostof the growth occurred in the 1970s in the irrigatedareas and in the 1980s in the rainfed areas. The totalfactor productivity actually declined in both types ofareas in the 1990s. Land productivity had grown at ahigher rate in irrigated areas than in rainfed areas.During 1970–73 to1992–95, the productivity of majordryland crops also increased. For example, the averageyield of coarse cereals has increased by 65.4% and thatof oilseeds by 57% (Kanwar 1999).

5.3.2 Low level of adoption of newtechnologiesAlthough new dryland agricultural technologies areavailable with research centers, their adoption rate isvery low in most of the SAT areas. For example, in1997–98, the area under High Yielding Varieties(HYVs) in the SAT was nearly 32% as contrasted with42% in the non-SAT areas in India and the averagefertilizer use was 97 kg ha-1 vis-a-vis 117 kg ha-1 in non-SAT areas (ICRISAT database. Refer to Table 1.2).Consequently, there is a wide gap between the averageyields obtained in research plots and the farmers’fields. The gap is due mainly to the inability of farmersto buy new inputs, their risk aversion, lack of adequateextension and other support systems such as access tomarkets and institutional credit, and the risk involvedin the adoption of new technologies (Mahipal andRamachandran 1999). This is a serious constraint inincreasing yield, production and income of the SATfarmers. The gap needs to be bridged throughappropriate interventions including establishing an

35

effective interface between research, extension,support systems and farmers and providing easy accessto institutional credit and crop insurance.

5.3.3 Increasing importance oflivestock and feed grainsLivestock rearing has been an integral part of farmingsystems in India since time immemorial, especially inSAT, where crop production is highly risky because ofits dependence on uncertain rainfall. So far, livestockresearch and development has not received as muchattention as it deserves. Now that the demand forlivestock products in India is increasing at a faster ratethan the demand for food grains, livestock productionwill become more financially attractive than in thepast. Besides, the growth in the livestock sector willcreate growth in the derived demand for feed grainsand stover of crops such as sorghum and millet (Kelleyand Parthasarathy 1996; Marshland and Parthasarathy1999). For example, in Sholapur market, the grain tostraw price ratio for sorghum declined from 6:1 to 3:1between early eighties and early nineties (Kelley et al.1993). There is hence a compelling case to shiftemphasis in genetic improvement of sorghum andmillet from food grain to feed grain traits and withemphasis on straw yield and quality. The InternationalCrops Research Institute for the Semi-Arid Tropics(ICRISAT) is already addressing projects related tofeed use of sorghum and millets as poultry feed, andimproving the fodder quality of groundnut, sorghum,and pearl millet funded by the Department forInternational Development (DFID) and theAustralian Center for International AgriculturalResearch (ACIAR). A bio-economic study of the valueand desirability of this shift is required. Increasedattention is needed on the integration of livestock inmixed crop-livestock systems, beyond the currentemphasis on improving stover quality.

5.3.4 Bridging the gap betweenresearch and developmentStrengthening the capacities of SAT farmers andnational research systems with the aid of informationtechnology will lead to significant rewards. This wouldhelp bridge the gap between the potential crop yieldsas demonstrated in research plots and those obtainedin farmers’ fields. ICRISAT can play a key facilitativerole in the process of information delivery/exchangeand training. Improved access to informationtechnology will also enable the SAT farmers to obtainreal-time information on markets, prices, weather, and

pest and disease epidemic forecasts. This can furtheropen new commercial opportunities and reduce theinherent risks of SAT agriculture.

5.4 Economic problems

5.4.1 Lack of financial viability

Given the low productivity and fragile nature ofnatural resources in SAT, it would be advisable toimprove the financial viability of SAT agriculturethrough promoting the production of high value crops,having good domestic market and export potential.Examples of such crops include spices, medicinalherbs, fruits and vegetables.

5.4.2 Low public investment inagricultureThe slow pace of growth of agricultural economy andits impact can be partly explained by low investmentsin agriculture in SAT. Investment in agriculturewhether public or private is important part of growthprocess and, therefore, it is an important resource.The gross capital formation in agriculture has beendeclining in both public and private sectors over time.The other dimension of this issue is the low returnsfrom these investments, which is reflected in theanalysis of incremental capital output ratio (ICOR). Itis believed that the public investments in agriculturehad a very narrow focus mainly on irrigation but otherinfrastructure components such as roads, markets,rural electrification, etc, were neglected. Investmentin irrigation and infrastructure, particularly roads andmarkets, have higher marginal returns in the SAT thanin most of the irrigated areas (Fan et al. 2000).

5.5 Institutional/organizationalproblems

5.5.1 Lack of appropriate institutions

It is now clear that the emerging challenges inagriculture transformation would require newinstitutions with new orientations and functions tocope up with the fast changing requirements of theagricultural production system. There are severalforms of institutions extant in SAT. They include SelfHelp Groups (SHGs), Panchayati Raj Institutions(PRIs), and a host of other public and privateagencies.

Emerging changes and issues in SAT agriculture

36

Overcoming Poverty in Rural India

this purpose should be reviewed, and appropriateinstitutional structures, owned and controlled by farmfamilies, should be promoted.

5.5.2 Lack of community participationThe community participation is essential for thesustainability of development programs, as it helps inreducing the monitoring and other recurring costs,besides ensuring appropriate decisions. However, theparticipation of communities cannot be taken forgranted. It will come about and will be sustainable onlyif appropriate institutions such as empoweredpanchayats are built or allowed to exist and function.There is a lack of community participation in makingdecisions that affect the community and their livelihoodsin India in general and the SAT in particular. This has hadan adverse effect on the implementation of developmentprograms, reducing their effectiveness and impacts.

For enlisting community participation in agriculturaldevelopment projects, it would be necessary todecentralize decision making and empowering localinstitutions through training of their personnel andthrough technical and financial support. Besides,communities also need to be empowered to achievebetter implementation of agricultural developmentprograms. Greater involvement of women in decision-making in local institutions dealing with naturalresource management should also be deliberatelysought and encouraged.

5.5.3 Inappropriate property rightsand tenures in land and waterIn many of the SAT areas, land tenure and land leasingsystems are not yet reformed with the result thatthere are no incentives to the tiller to invest in landdevelopment including soil and water conservation.Given the widespread practice of land leasing on oralterms in SAT, as also in the non-SAT areas in India, it isnecessary to legally legitimatize it. This would not onlyprovide the needed incentives to the tiller to use newtechnologies and investment in land development, butalso enable him to access institutional credit. Similarly,most of the problems relating to the use andmanagement of water such as overexploitation,depletion and degradation of water resources arisefrom the lack of well-defined property rights in water.In a nutshell, they suffer from what Hardin (1968)called “The Tragedy of the Commons” (Singh 1995b)

In our opinion, for efficient and equitable watermanagement, it is necessary to vest the ownershiprights in the State and usufruct rights in water users’

We would like to emphasize that SHGs, PRIs, andother institutions should form an interlinked chain tohave an institutional environment conducive forsustainable agricultural growth and development andconservation of environment. SHGs motivate andactivate the poor to participate in the processes ofagricultural growth and development. PRIs raise theirstatus in the local community and make them partnersin the local-level planning for growth, developmentand improvement of environment. Public and privateagencies of the desired type would enable the poor andthe backward to establish themselves in themainstream. It is this interlinked chain of institutions,which would help the SAT agriculture to step into themainstream with a fair measure of hope andconfidence. It would be necessary for the SATgovernments concerned to formulate a strategy forinstitutions building focused on two objectives: (1)facilitating the formation and growth of self-reliantinstitutions; and (2) acquiring adequate capabilities tointervene and manage crises beyond the capacity ofthe normal institutional mechanisms to cope with.

Yet another school of thought views theinstitutional sub-system in a broad perspectiveextending beyond the conventional concerns and therole of government-run organizations. The currentlyfashionable view is that the government is excessivelyinvolved, intervenes far too much and that tooinefficiently. The remedy, it is argued, would be totrim the government’s role and allow the privatesector to play a larger role. It is true that there aresubstantial inefficiencies in the functioning ofgovernment and parastatal agencies providing thenecessary support facilities and services. But it doesnot follow, as proponents of privatization assume, thatthese problems will be automatically and satisfactorilyresolved by handing these activities over to privateenterprises. While recognizing that there is no escapefrom the necessity of public systems, we need to focuson devising a better distribution of functions betweengovernmental and private organizations; examine thestructure and working of public systems (includingregulatory agencies) in key sectors; identify thereasons for their malfunctioning; and devise ways toenable and induce them to be more efficient andaccountable for performance.

We believe that ‘user-controlled and user-driveninstitutional structures’ characterized by lowtransaction costs are essential to provide the neededassistance in postharvest technology such as drying,storage, processing, and marketing. In this context, wewould suggest that the role of the Small Farmers Agri-business Consortium (SFAC) that was established for

37

organisations (WUOs), or the PRIs. This should bedone through the enactment of necessary legislationby the concerned state governments.

5.6 Administrative andmanagement problemsMost often policy makers presume that whatever theobjectives of agricultural development policy, theexisting administrative machinery can achieve them.This has not been borne out by India’s experience withthe implementation of various agricultural policies andprograms in the past. Furthermore, this contradicts awell-known management principle underlyingsuccessful development programs that for eachobjective, an appropriate organization structure shouldbe specified and that both the objectives and thestructure should be congruent with the prevailingsocio-economic and political environment.

The implementation of the new strategy foragricultural development would require thatagricultural development programs be professionallymanaged. For professionalizing the management ofagricultural development, we need appropriatelytrained agricultural development managers. Suchprofessionally trained managers with the requisitetraits, skills and commitment are a scarce commodityin India at present. A very large number of suchmanagers are required for managing agriculturaldevelopment programs and projects, rural institutionsand organizations, rural enterprises and resources.There is, therefore, an urgent need to establish manymore institutes of agricultural and rural developmentmanagement on the pattern of the Institute of RuralManagement, Anand. Appropriate programs andfacilities are also needed for imparting training to ruralproducers, rural women, rural bankers, agri-businessdealers, local officials, policy makers and planners, andpoliticians.

5.7 Public policy-relatedproblems

5.7.1 Food and nutrition insecurityIt is rightly stated that India is no longer facing foodinsecurity, but continues to suffer from nutritioninsecurity. This being the ground reality, nutritionalconsiderations should receive highest priority in foodproduction policies. This goal of nutrition and foodsecurity becomes all the more important consideringthe fact that the SAT area is one of the most food

insecure regions in India in terms of food availability,food access and food absorption. (Food InsecurityAtlas of Rural India, World Food Programme and MSSwaminathan Research Foundation, 2001).

Considering the above viewpoints, the orientationof this strategic goal should be one of enhancing theaccess to food for people who are food insecure, toachieve what is known as ‘human sustainability’. Inaddition, the effort should also be made to improvethe diets through nutritional orientation of the existingagricultural production system, eg, inclusion oflegumes in the cropping system. Further, one needs toconsider the emerging empirical evidences which nowsuggest that income in the hands of womencontributes much more to household food securityand child nutrition than income controlled by men.This will call for ways and means to channelizedevelopmental investments in women to improvefood security for the whole family. It is felt that ifwomen are left out, the progress of nutrition and foodsecurity programs cannot be ensured.

5.7.2 Lack of access to markets andmarketing facilitiesThe marketing of farm produce has been the mostneglected dimension of agricultural developmentstrategies followed in India in the past. It is high timethat it were accorded its right place in the futurestrategic action plans of the development of SATagriculture in line with the 2020 Vision statement. Inthe process of developing agricultural markets, enoughsafeguards need to be built to protect the interest ofsmall and marginal farmers and other weaker sections.

In the wake of new world trade regime ushered inby the World Trade Organisation (WTO), thegovernment is withdrawing from procurement, storage,and distribution of food grains. Simultaneously, we arealso witnessing the removal of restrictions on storage,sale and movement of agro-products, removal of exportcontrols, and expansion of futures and forward tradingin agro-products. How these changes will influencethe farm produce markets in the SAT is not yet clear.

However, be that as it may, the future financialviability of agriculture in the SAT will be greatlyinfluenced by the efficiency of farm productionsystems on one hand and efficiency of marketing, onthe other. The past experience suggests that the task isnot going to be easy. The issues, which need to beeffectively countered include• overcoming space- and time-related uncertainties

faced by farmers in general and small and marginalfarmers in particular;

Emerging changes and issues in SAT agriculture

38

Overcoming Poverty in Rural India

thin. Consequently, the impact of the interventions isnot visible. In view of this, there is an immediate needfor the state governments concerned to consolidatetheir manpower and other resources around selectedinterventions. This consolidation effort may bereviewed from time to time depending on theexperience gained.

It must be admitted that the support systemsextant in the SAT are weak and farmers face lots ofproblems in accessing their services. For example, theregulated markets, particularly those located in farflung areas, are thin with low trade volumes and fewbuyers and sellers. The market makers buy atrelatively low price and sell at higher price with largemargins. The markets are not properly integrated. Astudy on price spreads in metro markets sponsored bythe Planning Commission revealed that reasons forhigh spread between the wholesale and retail pricescould be the high concentration of trade with fewtraders, high transaction costs, weak infrastructureand information system and a large number ofintermediaries in between the wholesalers and theretailers.

5.7.4 Regional strategies anddecentralized planningThe SAT areas in India are highly heterogeneous interms of natural and human resource endowments,types of farming systems, levels of living, livelihoodpatterns and infrastructure. In view of this, no uniformR&D strategy would be appropriate for the SAT as a

• non-existence and/or inefficient markets foragricultural produce, particularly for perishablecommodities;

• structural changes in the marketing system such asprivatization and multi-agency competition;

• expanding market opportunities includingaccessing overseas markets, particularly forprogressive farmers growing high value, export-oriented commodities;

• use of latest information and communicationtechnology (ICT) to improve the integration andefficiency of agricultural markets; and

• marketing and extension support to help farmers to‘manage risk’ and the process of transition fromsubsistence agriculture to market-orientedcommercial agriculture.Research on reducing the high marketing and

transaction costs through innovative and effectivemarket linkages of producer and end users in the SATis likely to bring in good rewards than a focus ondeveloping new postharvest and processingtechnologies.

5.7.3 Weak support systemsFive major crops, viz, oilseeds, cotton, pulses, wheat,and sorghum account for nearly 75% of the grosscropped area (GCA) of SAT. The remainder of theagriculture is highly diversified and includes a widerange of crops. A glance through the schemes andprograms in operation in the SAT indicates that theinterventions in the SAT are spread too far and too

Table 5.1. Comparison of crop yields (kg ha-1) in SAT & Non-SAT regions of India.

SAT Non-SAT All India

Crops/Year 1970 1998 1970 1998 1970 1998

Rice 1659 1753 801 1871 1120 1880Wheat 947 1813 1442 2801 1310 2670Sorghum 250 933 1508 750 470 960Pearl Millet 517 928 487 693 620 790Maize 1074 1637 1025 1752 1280 1700Finger Millet 897 1463 703 1020 870 1350Barley 1084 1348 1040 2019 1090 1890Chickpea 549 788 723 796 660 810Pigeonpea 281 626 1304 817 710 750Groundnut 595 1166 1427 1089 830 1150Rape and Mustard 1479 926 355 832 590 1010Sugarcane1 6629 8029 4477 6211 4832 6652Cotton 80 223 374 305 110 270

1.Based on gur productionSource: ICRISAT database

39

whole. There is a need for differential strategies suitingto the differential characteristics of various areas. Forthis, a decentralized system of decision-making andplanning is required. The 73rd (Constitution)Amendment Act can facilitate this process if the SATstate governments are willing to devolve administrativeand financial powers to PRIs.

5.7.5 Mainstreaming gender inagricultural development strategyIncreased seasonal and permanent male migrationfrom rural to urban areas is leading to the feminizationof SAT agriculture in India as well as in Sub-SaharanAfrica (SSA). There have been several references tofeminization of agriculture in many poor areas ofAfrica and Asia. As part of the micro-level evidencethat ICRISAT is gathering in several watershedmanagement research locations, data were collected in2002 in six villages in Ranga Reddy district of AndhraPradesh, India. A census was undertaken before adetailed sample survey. Among the 825 householdscovered in the census, only 5% were registered aswomen-headed households. About 52% of thehousehold labor force is composed of male workers,while the balance of about 48% is from womenworkers. The detailed sample survey (of 120households) data shows other interesting results. Onan average, women contribute 70% of the 124 daysha-1 of labor required in crop production activities.About 50% of this labor is hired within the village,from which women labor constitutes about 93%. Thisindicates that rural agricultural labor markets arehighly dominated by women. The findings also showthat women provide a significantly high proportion oflabor into labor-intensive activities such astransplanting, weeding, harvesting, and threshing ofcrops. However, the average wage rates for womenworkers are less than half the rates for male workerson the farm (about Rs 18 day-1 vs Rs 40 day-1). Thisindicates the rigidities and institutional barriers thatexist within the rural labor markets, which do notrespond effectively to the growing role of women andtheir productivity in SAT agriculture.

This trend is accompanied by increasing laborscarcities. R&D institutions need to recognize theneed for labor and capital-saving technology optionsthat are purposefully designed to be gender-sensitive.Besides, it should also be recognized that the successof projects to foster sustainable agriculturaldevelopment largely depends on the extent to whichboth women and men participate in project design,planning, implementation and monitoring. There aremany ways of institutionalizing a gender-sensitiveapproach to sustainable development. The first andforemost requirement is the policy of equalinvolvement of both men and women in developmentprojects. Next, women should be assigned specificroles, responsibilities and rights in making variousdecisions rather than simply attending projectmeetings. Besides, the whole project team has to besensitized to gender issues if gender awareness is to beencouraged. The whole team also has to be trained andmade responsible for the implementation of thegender-sensitive approach. Finally, it is also importantto clearly define the objectives to be achieved andmeasures to be employed to achieve each of theobjectives, and develop quality and process-orientedindicators in order to monitor the progress of theprojects in achieving the intended objectives.

5.8 Concluding remarksThere are several biophysical, technological, economic,institutional/organizational, and public-policy-relatedconstraints that hold back the rate of agriculturalgrowth and development in SAT and therebycontribute to the perpetuation of poverty, hunger,malnutrition, and low levels of living in the region.These constraints must be overcome on the grounds ofefficiency, equity and sustainability of agriculturaldevelopment. India now has the technologies suitablefor overcoming, or reducing the severity of most ofthese constraints. What is needed is a pro-SAT publicpolicy as briefly presented in Chapter 6 that commitsenough resources to resolve the endemic problems ofSAT.

Emerging changes and issues in SAT agriculture

6. A Development Strategy for SAT Agriculture

policy, research, and extension support is needed tomake agriculture financially viable for the majority offarmers? What priority areas for future investmentswould increase the contribution of SAT agriculture inreducing poverty in a sustainable manner? Despite thecomplex relationship between agricultural researchand poverty reduction, future research strategy inagriculture should be responsive to the felt needs ofthe poor farmers and aim at generating yield-increasing, drought-resistant and labor-intensive tech-nologies suitable for agro-climatic conditions in whichthey live.

6.2 Priority thematic areas forpossible developmentinterventionsBased on our characterization of SAT, identification ofcritical issues in SAT agriculture, and experience withand lessons of current development interventions inSAT presented in Chapters 3,4 and 5, we haveidentified the following thematic areas for develop-ment intervention in SAT in future:

6.2.1 Water as an Entry PointAs we stated in Chapter 3, water scarcity has been themost critical constraint in SAT agriculture. The role ofwater as a catalyst of socio-economic development inSAT has now been amply demonstrated andestablished (Singh 1997). In the regression modelsestimated and reported in Chapter 3, we found thatlow irrigated areas in all zones have high incidence ofpoverty. Furthermore, we also found that irrigation isan important determinant of agricultural productivity,which in turn is negatively correlated with poverty. Inview of this, optimal harvesting of surface run off,recharge of groundwater, judicious use andmanagement of available water should be the coreelement of the future strategy. In a nutshell, thefollowing water-related interventions need to be madein SAT on a high priority basis:

Drought Proofing through Harvesting and Storageof Rain Water: As we stated earlier in this report, the

6.1 IntroductionThe poverty diagnostics and strategic assessmentsundertaken in Chapter 3, 4 and 5 addressing the issuesconfronting Semi-Arid Tropics (SAT) generated acomprehensive perspective from both micro andmacro levels which deepened our understanding of themulti-dimensional nature and dynamics of poverty. Alinkage between rural poverty and SAT agriculture isestablished and the determinants of poverty areidentified. This is mainly based on the analysis of theincidence and determinants of poverty in SAT, keyissues confronting SAT agriculture, livelihoodstrategies adopted by the SAT households, and theexperience with and lessons of past interventions. Thisunderstanding is valuable in identifying and designingappropriate interventions to help the poor achievepreferred livelihood outcomes, improve their access toproductive natural resources and technology, andevolve strategies from an investment angle, andenhance their standards of living.

The strategies proposed in this chapter are allsubsumed in the Sustainable Livelihoods Approach(SLA) that we briefly presented in Chapter 2. Itrecognizes that poverty alleviation projects should bemore people-centered, responsive and participatory,multi-level through partnerships, sustainable, anddynamic. It calls for a process approach in research andintervention implementation, which means updatingthe existing knowledge all the time and generating newknowledge about livelihoods. An importantmanifestation of poverty in SAT is having few or nooptions and economic opportunities for diversificationand improving one’s livelihood. Therefore, futurepolicies must try to increase the number of optionsavailable to the poor by improving access totechnology, appropriate crop and livestock varieties/species, information, encouraging mobility andpromoting markets, creating infrastructure and givingthe poor a voice.

The chapter contains set of recommendations onstrategies to increase the contribution of SATagriculture to reduction of poverty. The developmentof a strategy for SAT agriculture evolved byconsidering major questions: What is the future ofagriculture in less favorable regions such as SAT? Whatkind of agriculture would be sustainable? What kind of

40

41

most serious problem of SAT agriculture in India isdeficient rainfall and its highly uneven distributionover time and across space. India now has the requisitetechnology, manpower, and material resourcesavailable for solving this problem. Pioneering work indesigning and building cost-effective small check damsin project areas comprising over 600 semi-arid villagesin three western Indian states of Gujarat, MadhyaPradesh and Rajasthan (Singh and Gupta 1998) can beused as a model. This strategy is people-centered andcommunity-based and seems to be appropriate forreplication in SAT areas in India. The financialresources required could be mobilized from the on-going programs.

Recharging of Depleting Groundwater Aquifers andRegulation of Groundwater Extraction: Deficientrainfall together with the overuse of the groundwaterin many SAT areas has led to the lowering down ofwater table. This implies that groundwater extractionin those areas exceeds the natural recharge rate. Thishas not only increased the cost of water extractionfrom tube wells, but also has resulted in the drying upof shallow wells in the areas. In view of this, there is adire need of recharging groundwater aquifers andimprovement in the on-farm water management. Theextraction of groundwater needs to be regulated bythe communities concerned, or water users’associations, wherever they exist. Innovative methodshave been tried by the Non-GovernmentalOrganizations (NGOs) that are cost effective andsimple to construct and maintain. These are suitablefor replication in other SAT areas in India.

Adoption of Watershed Management Approach:Watershed management approach has been found tobe the most appropriate strategy for optimum andsustainable use of land, water and other resources inrainfed farming areas. There are many success storiesincluding those of Sukhomajri, Ralegan Siddhi, andPIDOW (Gulbarga), that document how the adoptionof watershed approach led to overall development ofthe watershed and restoration of its natural capital insemi-arid areas of India (Singh 1995a). Improvedwatershed development technologies are nowavailable with various research institutes in most statesin India. What is needed is an organizational structurefor compiling, collating, screening and releasing newtechnologies found suitable for adoption by farmers.(Singh, Research Paper, IRMA).

Pricing of Water and Power Reflecting theirOpportunity Costs: To promote sustainable use ofwater, it is necessary to do away with all kinds ofdistortions in the existing markets in a phased manner.

Pricing of water and electricity in India is a highly,politically sensitive issue and there is reluctance on thepart of state governments to increase water and powerprices for agriculture. However, there is a plenty ofevidence now available in India to show that waterusers in both rural and urban areas are willing to paymuch more for dependable, safe, timely, and adequatewater supplies and sanitation services. For example, ina sample survey of 100 farmers conducted in twopurposively selected villages in Kheda district ofGujarat, it was found that the sample farmers werewilling to pay 150 to 300% of the prevailing tariff fortimely and adequate supply of irrigation water (Singhand Ghatak 1995).

Improving Water Use Efficiency of IrrigatedAgriculture: It is well known that agriculturalproduction in the Green Revolution period had grownat a faster rate in areas with assured irrigation andwhere judicious use of both surface and groundwaterresources was practiced. On the other hand,agricultural productivity remained low in areas wherewater management was poor and SAT is one of thoseareas. The measures required to improve the wateruse efficiency in SAT include promotion throughsubsidies and extension education the use of watersaving micro irrigation technologies such as drips andsprinklers and organizing water users’ associations tomanage on-farm use of water and to levy and recoverwater charges.

Specification and Enforcement of Property Rights inWater: Most of the problems relating to the use andmanagement of water such as overexploitation,depletion and degradation of water resources arisefrom the lack of well-defined property rights in water.There are no explicit statements or laws, which clearlyrecognize and define property rights in either surfacewater or groundwater (Singh 2001).

In our opinion, for efficient and equitable watermanagement, it is necessary to create property rightsin water through water laws and vest the ownershiprights in the State and usufruct rights in water users’organisations (WUOs), or Panchayat Raj Institutions(PRIs).

6.2.2 Better Management ofWastelands and Common Pool LandsIn Chapter 3, we found that there is a positivecorrelation between the extent of fallow lands andpoverty. Most of the degraded/waste lands includingfallow lands in SAT could be brought under cultivationwith relatively low investment and could be used for

A development strategy for SAT agriculture

42

Overcoming Poverty in Rural India

agro-forestry, community/social forestry, and horti-cultural crops. The International Crops ResearchInstitute for the Semi-Arid Tropics (ICRISAT) hasalso developed improved technology designed tofacilitate double cropping in those regions of theIndian SAT, which are endowed with deep Vertisolsand having an average annual rainfall of over 750 mm.

6.2.3 Diversification with HigherFocus on Livestock DevelopmentAs we stated in Chapter 4, at the micro level,diversification of agriculture is both a coping strategyagainst risk and an income enhancing strategy subjectto available resources of land, labor and capital.

A recent study by Birthal and Parthasarathy Rao(2002) has shown that diversification towards highvalue crops reduces interpersonal disparities inincome. All the livestock products such as milk, meatetc, were identified as inequality reducing. Fruits andvegetables cultivation too reduces income inequalityalthough to a lesser extent compared to livestockproducts. On the other hand, crops such as oilseedsand pulses and commercial crops are inequality-increasing sources of income.

There is a need for the adoption of regionallydifferentiated strategies for increasing agriculturalproduction in SAT. This strategy aims at promoting thecultivation of those commodities which are specific tocertain geographical areas and have become popularfor their differentiated characteristics/quality traits/suitability for agro-processing. Some of these cropsinclude medicinal and aromatic plants such as isabgoland essential oil grasses such as palmarosa, vetiver andlemon grass which are cultivated in many areas in SAT(Pratibha et al. 1999) and floriculture and organicfarming.

The crops mentioned above command a highpremium in the market and directly contribute toimproving the profitability of farming. Reputedorganizations of farmers or industry associations or agri-business firms engaged in trading and agro-processingmay be encouraged to take over the responsibility ofdeveloping these specific crops and areas by establishinglinkages between the farmers and markets.

6.2.4 Marketing and CommercialOrientation of AgricultureTo keep with the changing world trade regimecharacterized by globalization and commercializationof agriculture and changing food habits of people infavor of livestock products and fruits and vegetables,

SAT farmers will need to have market orientation inmaking decisions about crops they should grow. Infact, this is already happening in SAT (Gulati andKelley 1999).

As we saw in Chapter 3, access to good markets (asmeasured through market density), which can ensurefair price to the producer, is essential for increasing theproduction and profitability of agriculture in SAT. Theexperience gained in India and elsewhere suggests thatif agricultural marketing system is inadequatelydeveloped, efforts to increase agricultural productionare unlikely to succeed. By and large, quite a significantproportion of agricultural produce ranging from 30 to75% is still sold in villages to middlemen. In most ofthe cases, the producer does not get a fair price whenhe sells his produce in the village itself. There is a needfor launching a campaign using mass media informingfarmers about the prices prevailing in regulatedmarkets and the facilities available to them.

Contract Farming or vertical coordination of variousdegrees is emerging as an alternative to open markets.Under these arrangements, rather than selling theproduce on the commodity market, these transactionsrely less on prices and more on private contractnegotiations. It is believed that these alternativearrangements will give growers an opportunity to makerelatively more profits as compared to open markets.Such arrangements are gradually evolving in the countryincluding SAT. However, there is a need to sort outseveral issues including the mode of sharing of risks andprofits of the new market system among the playersfrom the traditional markets. Such exchangearrangements will have substantial impact on farmmarketing and production practices.

Regarding value addition, it is generally felt that thesmall size of farms does not permit economies of scale.Producing on small farms but processing on large scaleto reap the economies of scale can overcome this.Appropriate farmer groups or associations could tie upwith processing industries and thus share the benefitsof value addition. To overcome some of the problemsin the existing scenario, it is necessary that farmersintegrate themselves backward for credit and forwardfor value added processing. Institutional credit mayfree them from the hold of money lenders-traders andalso impart greater flexibility in marketing theirproduct.

6.2.5 Institutional Innovations

In the context of the proposed comprehensivedevelopment strategy for the SAT agriculture for thenext two decades, it would be necessary to identify

43

Normally, women’s rights to water and to ‘ownership’of water infrastructure is restricted because of theirlimited access to land rights or to their recognition as‘heads of households’ (Bina Agarwal’s “A field of myown”, Cambridge University Press). A “WaterCampaign” launched in 1995 in selected villages inGujarat by an NGO, Self-Employed Women’sAssociation (SEWA) is a good example of involvingwomen in decision making, raising their andcommunity’s awareness about water-related problems,and building and enhancing their capacity throughtraining and leadership development programs(IRMA-UNICEF 2000).

Using Mass Media to Create Awareness about theProblems of SAT Agriculture: Mass media could playan important role in generating mass awareness aboutthe impending water scarcity and droughts and theneed for water conservation in SAT. At present theelectronic and print media has not given a high priorityto educating and sensitizing the general public anddifferent stakeholders about the water scarcity and theneed for conservation of water resources. There is anurgent need to devise innovative ways as well to usethe indigenous methods such as folklore and folk songsfor the purpose. Programs such as “water campaign”,water day, and so on need to be used more widely tocover all parts of India and to generate public opinionon water related issues. There is a need for developinga mass communication strategy and a national campaignfor building public consensus on institutional andindividual actions with regard to improved SATagriculture and its management (IRMA-UNICEF 2000).

6.2.6 Building and Strengthening ofthe Basic Infrastructure

Agricultural productivity and production, and farmand non-farm incomes in SAT are constrained bydeficient infrastructure comprising roads, markets,hospitals, electricity, irrigation, means of transport,and communication. This imposes a severe constrainton increasing production. It is now widely acceptedthat the marginal returns to investment ininfrastructure, especially roads, irrigation, electricityand education, in relatively less developed rainfedareas such as SAT are higher than in relatively moredeveloped irrigated areas (Fan et al. 2000; Ryan andSpencer 2001). In view of this and given the higherincidence and severity of poverty in SAT than in thenon-SAT areas in India, higher investment in basicinfrastructure is justified on both efficiency and equitygrounds.

the best practices and institutional/organizationalinnovations that are already being used by progressivefarmers and NGOs in SAT. Some of those innovationsare briefly described below:

Demand-Driven Agricultural Extension System: Themain objective of the agricultural extension systemshould be to cater most cost-effectively to theemerging needs and demands of the SAT farmers,especially the small and marginal farmers. Theadoption of the Training and Visit (T&V) systemduring the 1980s was a bold step. A similar innovativemeasure is required to be taken now. The UnionMinistry of Agriculture needs to re-orient the T&Vsystem and the research and extension institutions tocomplement this. Every state needs to revamp itsextension system. Training institutes are in dire needof rehabilitation, whether it is physical infrastructure,training methodologies or trainers. The Governmentof India (GOI) should put this in the center of theiragenda. While an upgrade can be according to a planand investments need-based, the Union Ministry ofAgriculture should not deter from providing supportto all the three components cited above.Mainstreaming of gender concerns should also beplaced at the heart of this revamp strategy.

Professionalization of Management: There is a needfor professionalization of agricultural developmentmanagement to cope with the rapidly changingnational and international economic and politicalenvironment. This could be achieved only slowly overa period of time by inducting professionally-trainedagricultural development managers at all levels in thehierarchy of development administration. Reorienta-tion of the existing staff through short-term trainingprograms in agricultural development management ingood institutes of rural management is also necessary.

Securing Women’s Participation: Women areimportant stakeholders in SAT agriculture. They areresponsible for fetching water for meeting theirhousehold requirements, collating fuel and fodder,tending animals, and performing various farmoperations except ploughing. Therefore, they shouldbe involved in making decisions that affect them.NGOs and community organizations should facilitatewomen’s participation in decision-making at thecommunity level by addressing structural constraints,ensure that women’s participation does not increasetheir already heavy workload, and train women andmen and build their capacity.

The government agencies should create andenforce women’s rights to land and water eitherthrough legal or new institutional mechanisms.

A development strategy for SAT agriculture

44

Overcoming Poverty in Rural India

6.2.7 Re-orienting Public PoliciesNow, we briefly describe a few of the policies andprograms that need to be refocused and streamlined tobe meaningful to the SAT farmers, especially as theyrelate to key factors (identified in Chapters 3, 4 and 5)constraining agricultural productivity and hencepoverty reduction.

Ensuring Food Security through Provision of Wage-paid Employment: This is the most important elementof any pragmatic strategy of management of SATagriculture. In Chapter 4, we highlighted the need tostabilize food grains prices as a means of reducing theimpact of income poverty. For poor families,household food security can be a threat duringdroughts. It can be strengthened through a moreeffective Public Distribution System (PDS),uninterrupted and flexible supplementary nutritionthrough Intensive Child Development Schemes(ICDS) and creation of food grain banks which mayalso be linked with food for work scheme.

A weekly system of rationing under PDS will bemore beneficial as really poor families cannot afford topurchase their whole entitlement in a matter of fewdays. Another way of improving the performance ofPDS is to raise public awareness about the program.

The role of women in the context of food securityalso needs to be understood. At various stages ofhousehold food production, women play an importantrole. But they are worst sufferers when it comes tofood scarcity. The organized women’s self-help groupshave a better capacity to manage household income,savings and credits. There is a lot of evidence availableto suggest that if appropriate training is provided,women’s self-help groups could be a better tool toreduce household food insecurity even for those whoare living in abject poverty.

Higher Public Investment in Infrastructure andTechnology: As discussed earlier in Chapter 5, SATagriculture is characterized by low levels of input useand low productivity levels. The average quantity offertilizers used and the proportion of cropped areaunder High Yielding Variety (HYV) in the SAT areasare much lower than those in the non-SAT areas. Toget out of this syndrome, we need to step up the levelof both public and private investment in infrastructureand technologies. Fan et al. (2000) have shown thatthe marginal returns to investment in infrastructureand technology in SAT areas are higher than those inirrigated areas. Further, on equity grounds also,farmers in SAT deserve a better treatment now interms of allocation of public funds than what has beenmeted out to them in the past. In the past, farmers in

irrigated areas have received better financial support inthe form of subsidized institutional credit, subsidies,and public investment in irrigation and infrastructure,perhaps at the cost of their counterparts in SAT.Investment in rural infrastructure particularly willhave a direct impact on food security through income,employment and wage impacts.

Higher Inflow of Institutional Credit to SATAgriculture: The per capita amount of institutionalcredit provided to farmers in the SAT is markedlylower than that in the non-SAT areas. This is both anti-equity and anti-efficiency, given the higher incidenceand severity of poverty and the higher returns toinvestment in the SAT agriculture. Besides, the termsand conditions of the credit provided are also notappropriate under the ecological and socio-culturalenvironment obtaining in SAT. It has been observed inchapter 4 that dryland agriculture is profitable over aperiod of 3 to 5 years even though in any one year itmay be a losing concern. In view of this, a new(cyclical) credit policy is required so as to meet the fullcredit requirements of the dryland farmer over theperiod of 3 to 5 years even when he becomes adefaulter in one or more than one year.

Covering More Households under Crop andLivestock Insurance: With the cost of cultivationgoing up and given the risk and uncertainty involved inagriculture, specially in SAT, every farmer is concernedabout the investment he makes and the returns heexpects for his and his family’s labor (as referred inchapter 4 under discussion in drought). It is indeed amatter of satisfaction that the Union Ministry ofAgriculture has already launched the NationalAgriculture Insurance Scheme. While the imple-mentation aspects and difficulties therein are a subjectmatter for detailed discussion, we would like to offer asuggestion that its coverage should be extended to allthe farmers in SAT.

Cover More Crops under the Minimum SupportPrices Scheme: Rainfed crops suffered substantialdiscrimination in the government’s procurement andpublic distribution policies. Although minimumsupport prices are announced for rainfed crops as well,they are seldom backed by procurement operations.Farmers who have grown these crops often incur losesbecause of low market prices. Many have shifted theircropping patterns by reducing the areas allocated tocoarse cereals. The PDS and the heavily subsidizedrice and wheat have further eroded thecompetitiveness of rainfed crops and altered marketprice ratios. Substituting the PDS with a food stampssystem leaves beneficiaries the option of buying grains

45

of their choice. In Andhra Pradesh, the use of couponsto distribute kerosene has benefited low-incomeconsumers. Similarly, food stamps worth Rs 150 permonth (equivalent to the current level of subsidy) maybe given to households below the poverty line toenable them to buy grains of their choice. This willserve two purposes – the benefit of food subsidy willbe spread over all grains and the disadvantages thatproducers of coarse grains face will be partially offset.Unless these policy initiatives to reverse the policybias are taken up vigorously, rainfed crops and farmersgrowing them may be marginalized further, forcingthem to seek livelihood options outside agriculture.

Rationalize Subsidies on Agricultural Inputs:Fertilizers, irrigation water and power (electricity) arethree of the farm inputs that are heavily subsidized atpresent in Indian agriculture. Irrigation subsidies arejustified, as it is unethical to penalize irrigators bycharging them high water rates based on real resourcecosts, which are abnormally high because of thewidespread inefficiency and mismanagement inIrrigation Departments. What is needed, in ouropinion, is both an improvement in efficiency andmanagement of Irrigation Departments andrationalization of the water pricing policy. It is also truethat quite a lot of so-called farm subsidies go to theprivate industries and public undertakingsmanufacturing farm inputs to cover up theirinefficiency and mismanagement. The existing policyof subsidies on agricultural inputs needs to bereviewed and its direct and indirect impacts ondifferent categories of farmers carefully assessed.There is, therefore, need to streamline the deliverysystem and to ensure that the benefits from subsidiesare widely and equitably distributed. To attain thisobjective, it is imperative that farmers in SAT are givena high priority in allocation of funds for subsidies onfarm inputs.

Facilitating Migration: As we discussed in Chapter 4,in SAT, seasonal, semi-permanent and permanentmigration is a predominant coping strategy adopted bythe poor to get out of the poverty trap. As a matter offact, an informal market for migrant labor hasdeveloped over time in India. Such informal marketsplay an important role in balancing the regional supplyand demand of casual labor in India.

Although the informal migrant labor markets inIndia are, by and large, working well, we believe thattheir efficiency could be improved and many morepoor could benefit from them, if an institutionalizedsystem of collection and dissemination of informationabout the supply, demand, and wage rates is put inplace at the district level in selected areas within SAT

and other poor areas of India. We would recommendthat the state governments must intervene moreeffectively in labor markets and ensure that the wagerates are fair and reasonable and that there is noexploitation of labor through such practices as bondedlabor, attached labor, and so on. This should be donethrough the involvement of some reputed NGOshaving a track record of good social service andcommitted to serving the cause of the poor labor.

Coping with the Challenges of Globalization andMarginalization: The increasing strides towardsglobalization of markets through domestic marketreforms that encourage integration and liberalizationof import and export markets, production efficiencyand competitiveness of agricultural products withinthe domestic market and international markets isbecoming an important policy issue in the agriculturalsector. Considering agriculture’s role as a means oflivelihoods for millions of poor people in SAT,enhancing its competitiveness through cutting averagecosts of production is critical for the survival of manysmallholder farmers.

Past empirical evidence in agricultural technologydevelopment and infrastructural investments in SouthAsia lends support to this process of marginalization inless-favored regions, especially the rainfed SAT. In thecase of India, Fan and Hazell (2000) show thatadoption of improved varieties, road density, marketaccess (number of rural markets per 1000 sq km), andintensity of fertilizer use are consistently lower inrainfed than in more-favored irrigated districts. Thehigh transaction costs and low productivity ofagriculture in the rainfed SAT will affect the relativecompetitiveness of smallholder crop-livestockproduction activities in these areas. It will alsoinfluence farm-household decision behavior in termsof crop and technology choice and ability to hedge risk,both from the market and from the adversebiophysical environment.

6.3 Priority geographical areasfor possible interventionsIdeally, and particularly on equity grounds, no area inSAT should be left out in the process of development.But, given the vast extent of the SAT areas in India andthe limited financial, human, and material resourcesavailable for their development, it is but imperative toprioritize the SAT areas for development interven-tions. The priority could be determined on the basis ofthe incidence and severity of poverty, and the potentialfor agricultural development. The highest prioritymust be given to those areas, which have the highest

A development strategy for SAT agriculture

46

Overcoming Poverty in Rural India

incidence and severity of poverty and where there ismedium to high potential for agricultural developmentin the short run. For areas where the potential foragricultural development is low, a separate strategyfocusing on non-agricultural avenues of incomeincluding migration should be developed.

Based on these two criteria, and keeping in view theresults of the regression analysis presented in Chapter3, we have identified the following developmentinterventions suited to the natural and human resourcesendowment and agro-climatic conditions obtaining inSAT (Table 6.1). The priority geographical areas forintervention are the low irrigated SAT areas coveringMaharashtra, Madhya Pradesh, Gujarat, Karnataka,Andhra Pradesh, Bihar and Rajasthan. Next in priority,are the medium irrigated areas covering parts ofRajasthan, Madhya Pradesh, Gujarat, Karnataka,Andhra Pradesh and Bihar. Even the high irrigatedareas of SAT spanning over Tamil Nadu, AndhraPradesh, Bihar, and Gujarat, also deserve attention ofpolicy makers in view of high levels of povertyprevalent there. Depending upon the availability ofresources, as many of the districts from these areas aspossible could be selected for intervention.

6.4 Priority areas for futureresearchAgricultural research needs to be farmer-oriented anddemand-driven so that it can generate profitabletechnologies suitable for various agro-climatic and

socio-cultural conditions obtaining in SAT. For this, itwill be necessary to follow both the land-to-lab andlab-to-land approaches; the former for identificationof farmers’ problems, and the latter for exchange ofnew technologies. Given the increasing importance oflivestock in the sources of livelihood and as a means ofminimizing weather-induced risks in SAT (see chapter4), there is also a need for focusing more efforts onlivestock development, improvement of pastures andgrazing lands, and on increasing forage yields of variousannual and perennial (tree) crops, particularly in SAT.The criteria used to evaluate the appropriateness ofnew technologies should include, besides grain yield,fodder yield of crops and trees.

Agricultural research should primarily focus onbringing about substantial improvement inproductivity and bridging the yield gap between theresearch stations and the farmers’ fields. As alreadystated, the radical improvements in productivity arepossible through new bio-technological research.Therefore, in the long run, there is a need to supportbio-technological research and place greater emphasison adaptive research in the short-run.

In view of the declining public support foragricultural research, it is important for thegovernments in the SAT areas to work in partnershipwith the private sector. The trends world over indicatethat most of the investments in biotechnology aremade by the private sector. Therefore, the stategovernments concerned need to find ways and meansto collaborate with the private sector in an

Table 6.1. High priority SAT areas for development interventions.

SAT with SAT with SAT withS.No. Type of intervention low irrigation medium irrigation high irrigation

1. Watershed management andsupplemental irrigation

2. Reclamation and better use ofwaste and fallow lands

3. Wider diffusion of new drylandtechnologies

4. Diversification with higher priorityto livestock development

5. Better targeted wage-paid employment

6. Education and training in new skills

7. Revamped and better targeted PDS

8. Strengthening of support systemand basic infrastructure

9. Facilitate migration X

Note: The number of tick marks (P) in Table 6.1 indicates the priority: 3 means high priority, 2 medium and 1 low priority. (X) means no priority.

47

environment where issues relating to intellectualproperty rights are also important. However, it mustbe remembered that the private sector investment canonly be a complement, and not a substitute for publicsector investment in research.

While determining high priority areas for research inthe context of SAT, it should also be kept in mind thatthe scope for targeting benefits to the poor throughreallocation of research resources among commoditiesis limited. Hence, enhancing the efficiency andeffectiveness of research systems in promoting broad-based technical change and its widespread diffusionshould be emphasized more than efforts to targetpoverty directly (Byerlee 2000). However, we wouldlike to assert that, despite the complex relationbetween agricultural research and poverty reduction,future research strategy should be responsive to thefelt needs of the SAT farmers and aim at generatingyield-increasing, drought-resistant and labor-intensivetechnologies suitable for agro-climatic conditionsextant in SAT (see also Otsuka, Keijiro, 2000). Besides,it is now high time that more research resources wereallocated to multidisciplinary problem-solving appliedresearch involving, among others, social sciences.

In setting priorities for research, it is necessary toinvolve the target users in the process. Participatorypriority setting can lead to different diagnoses andrecommendations for future agricultural research andproject implementation. One example of this is thesynthesis of a number of participatory surveys andbeneficiary assessments from Zambia, which showshow farmers’ own perceptions and priorities regardingconstraints to production and the quality ofagricultural services went beyond just resourceendowments and allocation (climatic andenvironmental factors, access to land, labor, skills,draft power) to the quality of public and privateagricultural services (Francis et al. 1997) This processhelped to identify the livelihood diversificationstrategies including changing farming practices, morereliance on non-farm sources of income, and modifiedpatterns of exchange and consumption.

Based on our analysis of the problems andprospects of the SAT agriculture in India attempted inthis report, and the results of a consultation held witha sample of stakeholders as reported by Ryan andSpencer (2001), we have identified the following highpriority areas for research in future:1. Breeding of water-efficient drought-tolerant

varieties of rainfed crops.2. Cost-effective methods and techniques of

integrated watershed development andmanagement.

3. Techno-economics of water saving micro irrigationtechnologies and determinants of their adoption inSAT..

4. Alternatives for making the typical farm familyfinancially viable through optimum mix of farmand non-farm enterprises.

5. A study of dynamics of rural poverty andlivelihood security in SAT.

6. A study of extent, determinants andconsequences of land degradation.

7. Mainstreaming of gender in SAT agriculturaldevelopment strategies.

8. Economics of postharvest technologies in SAT.9. Commercialization and globalization of

agriculture and its impacts on SAT agriculture.10. Strategies for promoting the use of modern

information and communication technologies inSAT agriculture.

11. Institutional innovations for reducing the hightransaction costs and marketing costs.

12. A critical appraisal of major public policyinterventions in SAT agriculture.

13. A study of Dynamics and determinants of rural-urban migration and its impacts.

14. A bio-economic study of the value and desirabilityof a shift in the focus of research fromimprovement of sorghum and millets as foodgrains to their improvement as feed grains.

6.5 Building partnerships andlinkagesThere are numerous governmental and non-governmental, and international and nationalorganizations and agencies directly or indirectlyinvolved in the development of SAT. At present thereis no institutional mechanism to coordinate theiractivities. This results in avoidable overlaps in theirfunctions and thereby wastage of scarce financial,human, and material resources. Besides, due to lack ofcommunication and coordination, many of theinitiatives are duplicated. In view of this, we suggestthat an institutional mechanism in the form of a formalautonomous organization, which may be named IndiaSAT Development Authority (INSATDA), may becreated at the national level, with state levelconstituent organizations. INSATDA may be assignedthe main responsibility to network, coordinate, andfacilitate the activities of all the GovernmentalOrganizations (GOs) and NGOs involved in the SATdevelopment. In particular, there is a dire need foreffective cooperation and collaboration among various

A development strategy for SAT agriculture

48

Overcoming Poverty in Rural India

Besides the empowerment of the SAT dwellers, it isalso equally important to build the capacity of the SATinstitutions and enable institutional learning andinnovations. It is now widely accepted that institutionsand organizations play an important role in the processof development. Organizations may affect agriculturaland rural development in many different ways includingprovision of production inputs and services, reductionof transaction costs, enhancement of bargaining powerof rural producers vis-a-vis those to whom they sell theirproduce and from whom they buy production inputsand services, influencing investments and savings andbringing the two together, and so on.

6.7 Concluding remarksGiven the fragile natural resource base, under-developed human resources, weak institutions andorganizations, and the inadequacy and poor quality ofbasic infrastructure in SAT, the Green Revolutionstrategy will need to be suitably modified to beapplicable to SAT. In fact, a completely different set oftechnologies, institutional arrangements and policies isrequired. One of the lessons of the experience ofGreen Revolution was that its benefits did not reachthe poor people and the backward areas. In view ofthis, we will need to better target the poor and therelatively more backward areas in SAT throughinstitutional and other innovations.

Given the serious and persistent problems of waterscarcity and droughts in SAT in India, using water as anentry point and as a catalyst of development shouldreceive the highest priority. The major areas of interven-tion include harvesting, storage and conservation of rainwater, recharging of groundwater aquifers, ensuringefficient use of water, and improving the design andimplementation of drought relief programs.

The mission of ICRISAT should be to direct andfacilitate the SAT agriculture to emerge as a market-oriented, commercially viable, and ecologicallysustainable means of producing food, fiber, rawmaterials and other commodities such that farmersfind it profitable and respectable to do so.

A broad vision for SAT agriculture is to reducepoverty, hunger, and malnutrition, and ensuresustainable livelihoods for everyone. This vision couldbe actualized through a multi-pronged strategy mainlycomprising (1) water as a catalyst of development;(2) diversification and selective specialization;(3) marketing and commercial orientation; (4)institutional innovations; (5) building and strengtheningof the basic infrastructure; (6) better targeting ofdevelopment interventions to the most needy people

State Agricultural Universities (SAUs), institutes ofIndian Council of Agricultural Research (ICAR), andthe Consultative Group on International AgriculturalResearch (CGIAR) institutes for deriving full benefitsfrom the complementarities and synergy that existamong their functions (Paroda 2001).

6.6 Capacity building anddevelopmentGiven the high incidence and severity of poverty, highilliteracy rate, lack of entitlements and the lack ofbasic infrastructure in SAT, most of the people livingthere have very few skills useful for non-farmavocations and they are vulnerable physically,economically, socially, and politically. They need to beempowered and their capacity built througheducation, training, provision of technical information,and institutional credit in order to enable them toparticipate in and contribute to the mainstreameconomic, social, and political activities. In ouropinion, the underdevelopment of human resources inSAT is both a cause and a consequence of theunderdevelopment of SAT. So, development of humanresources through empowerment seems to us a potentinstrument of initiating and sustaining the process ofdevelopment in SAT.

Training of farmers, especially small and marginalfarmers, women, rural youth, rural bankers, suppliersof production inputs and services, local officials, andpoliticians is necessary for agricultural development.Agricultural orientation at primary and secondarylevels and organization of vocational and technicaltraining in agriculture and allied activities forintermediate level workers are necessary for cateringto the technical manpower requirements of rural enter-prises and institutions at the village and block levels.

Although most agricultural and rural developmentprograms provide for the training of their beneficiariesand functionaries at all levels, the trainers, the trainingcurricula, materials, and facilities available are not upto the desired mark and, therefore, there is a need forimprovement in all these aspects of training. So far,not much attention has been given to the training ofpoliticians in agricultural and rural development.Properly trained and oriented, motivated, and well-informed politicians can be of great help in promotingthe cause of agricultural and rural development andthe opposite of it can be a great obstacle to ruraldevelopment. There is a need to develop appropriatecurricula and facilities for training of politicians at alllevels. In many cases, training of trainers would also benecessary.

49

and the most backward geographical areas within SATas well as ensuring the poor to access resources,institutions, technology and markets via build up ofsocial capital and empowerment; (7) focusing researchon the most relevant and salient problems of SAT; and(8) building partnerships and linkages. This strategydeserves to be pilot-tested in selected SAT areasinvolving both governmental and non-governmentalorganizations engaged in and committed to promotingsustainable livelihoods in SAT.

There is a need to prepare a National Advocacyplan for action at village, block, district, state and

national levels. The plan should aim at creatingawareness among people in SAT includingpoliticians about the need to manage their naturalresources, especially land and water resourcesjudiciously, mobilize people’s energies andresources and catalyze action. Through creating anawareness among people in general and primarystakeholders in particular about the extent andseverity of poverty and insecurity of livelihoods andchanging their attitude towards SAT agriculture,many of the current problems could be resolvedeffectively.

A development strategy for SAT agriculture

Bibliography

Birthal PS and Parthasarathy Rao P. 2002. TechnologiesFor Sustainable Livestock Production In India. Proceedings ofa workshop on documentation, adoption and impact oflivestock technologies in India, 20–21 January 2001,ICRISAT, Patancheru, Andhra Pradesh, India (Birthal PSand Parthasarathy Rao P, eds.). Patancheru 502 324, AndhraPradesh, India: International Crops Research Institute forthe Semi-Arid Tropics; and New Delhi, India: NationalCentre for Agricultural Economics and Policy Research.

Birthal PS, Joshi PK and Parthasarathy Rao P. 2002.Vertical integration in agriculture sector studies: dairy,poultry and fruits and vegetables. Paper presented at theworkshop on agricultural diversification in South Asia,jointly organized by RGOB, MoA, ICAR and IFPRI, Paro,Bhutan, 21–23 November 2002.

Bromley DW and Cernea MM. 1989. The management ofcommon property natural resources: some commonconceptual and operational fallacies. World Bank DiscussionPaper 57. Washington DC, USA: World Bank.

Buvinic M and Gupta GR. 1994. Targeting poor women-headed households and women-maintained families indeveloping countries: views on a policy dilemma. Mimeo,Washington DC: International Centre for Research onWomen.

Carney D (ed.). 1998. Sustainable rural livelihoods: Whatcontribution can we make? London: Department forInternational Development (DFID).

Chambers R and Conway G. 1992. Sustainable rurallivelihoods: practical concepts for the 21st century. IDSDiscussion Paper 296. Brighton: IDS.

Chambers R, Pacey A and Thrupp LA. 1989. Farmer first:Farmer innovation and agricultural research. London:Intermediate Technology Publications.

Chen MA. 1991. Coping with seasonality and drought.New Delhi, India: Sage Publications.

Coleman JS. 1990. Foundations of Social Theory. London:Belknap Press of Harvard University Press.

CRIDA (Central Research Institute for DrylandAgriculture). 1997. Vision-2020. CRIDA Perspective Plan.Central Research Institute for Dryland Agriculture, IndianCouncil for Agricultural Research, Hyderabad.

Davies S. 1996. Adaptable Livelihoods: Coping with foodinsecurity in the Malian Sahel. London: Macmillan.

Deaton A and Dreze J. 2002. Poverty and Inequality inIndia: A re-examination. Working Paper No. 107. NewDelhi, India: Centre for Development Economics, DelhiSchool of Economics.

Deshingkar P and Start D. 2003. Seasonal Migration forLivelihoods in India: Coping, Accumulation and Exclusion.Working Paper 220. London, UK: Overseas DevelopmentInstitute.

Adato M and Meinzen-Dick R. 2002. Assessing TheImpact of Agricultural Research on Poverty Using TheSustainable Livelihoods Framework. Food Consumptionand Nutrition Division (FCND). Discussion Paper 128/Environment and Production Technology Division (EPTD)Discussion Paper 89. Washington DC, USA: InternationalFood Policy Research Institute (IFPRI).

Agarwal B. 1989. Social security and the family in ruralIndia: coping with seasonality and calamity. DEP no 21.Suntory-Toyota International Centre for Economics andRelated Disciplines, London, U.K.

Arnold JEM. 1990. Common property management andsustainable development in India, Working paper 9. Forestryfor sustainable development program. USA: University ofMinnesota.

Atkinson AB. 1987. On the measurement of poverty.Econometrica 55:749–764.

Ballabh V. 2002. Emerging Water Crisis and PoliticalEconomy of Irrigation Reforms in India. Paper presented ata workshop on Asian Irrigation in Transition Responding tothe Challenges Ahead. Asian Institute of Technology,Bangkok, Thailand, 22–23 April 2002.

Ballabh V and Choudhary K. 1999. Small WaterHarvesting Structures: Technology, Institution andManagement. Workshop Report No. 15, Institute of RuralManagement, Anand

Ballabh V and Shah T. 1989. Efficiency and Equity inGroundwater Use and Management. Workshop Report No.3. Institute of Rural Management, Anand.

Ballabh V and Singh K. 1997. Competing Demands forWater in Sabarmati Basin: Potential and Future Conflicts.Paper presented at the IDPAD seminar, Amersfoort: TheNetherlands, 11–17 October 1997.

Ballabh V and Ujjawal P. 2000. India’s Water Crisis andInstitutional Challenges: An Overview. Paper presented atthe 8th Beinnial Conference of the IASCP, IndianaUniversity, Bloomington, 31 May – 4 June 2004.

Balooni and Singh K. 2001. Tree Plantations for Restorationof Degraded Lands and Greening of India: A Case Study ofTree Growers’ Cooperatives. Natural Resources Forum: AUnited Nations Journal 25(1).

Bauman P. 2001. Sustainable livelihoods and politicalcapital: Arguments and evidence from decentralisation andnatural resource management in India. Working Paper 136.London: Overseas Development Institute.

Bhardan PK. 1989. The New Institutional Economics andDevelopment Theory: A Brief Critical Assessment. WorldDevelopment 17(9):1389–1395.

50

51

ICRISAT (International Crops Research Institute for theSemi-Arid Tropics). 1998. Sustainable rainfed agricultureresearch and development project: database development,typology construction, and economic policy analysis,Module 1. Submitted to Indian council of AgriculturalResearch (ICAR) and Department of Agricultural Researchand Education (DARE).

ICRISAT–ICAR (International Crops Research Institutefor the Semi-Arid Tropics and Indian Council forAgricultural Research). 1999. Typology construction, andeconomic policy analysis. Sustainable rainfed agricultureresearch and development project (Module 1): Jointlypublished by ICRISAT and ICAR.137 pp.

IRMA–UNICEF (Institute of Rural Management, Anandand United Nations Children’s Fund). 2000. White Paperon Water in Gujarat, Institute of Rural Management, Anandand United Nations Children’s Fund, Gandhinagar, Gujarat.

Jackson C. 1995. Rescuing gender from the poverty trap.Gender Analysis in Development Series, 10. Norwich:University of East Anglia School of Development Studies.

Jodha NS. 1996. Common property resources and the ruralpoor in dry regions of India. Economic and Political Weekly,21.

Jodha NS. 1988. Poverty debate in India: a minority view,Economic and Political Weekly, Special Number, November1988, 2421-2428.

Kabeer N. 1995. Targeting women or transforminginstitutions? Policy lessons from NGO anti-poverty efforts.Development in Practice, 5, 2. Oxford: Oxfam.

Kanwar JS. 1999. Need for a Future Outlook and Mandatefor Dryland Agriculture in India. In Fifty Years of DrylandAgricultural Research in India, Central Research Institutefor Dryland Agriculture (CRIDA), Hyderabad.

Kelley TG, Jayawant M and Parthasarathy Rao P. 1997.Rainfed Agriculture typology in India. Economic andPolitical Weekly 32(26):A68–A70.

Kelley TG and Parthasarathy Rao P. 1993. Production andresearch environment for sorghum and millet in Asia, pp.93-124 in Sorghum and Millets Commodity and ResearchEnvironments, Byth D.E. (ed.). ICRISAT, Patancheru.

Kelley TG and Parthasarathy Rao P. 1995. Marginalenvironments and the poor: evidence from India, Economicand Political Weekly 33(40).

Kelley TG and Parthasarathy Rao P. 1996. Availability andrequirement of different sources of livestock feed in Indiawith special reference to sorghum and millet straw. Pages53–65 in Global Agenda for Livestock ResearchProceedings of the Consultation for the South Asia Region,8-6 June 1995, ICRISAT Asia Center, Patancheru. ILRI,Kenya.

Kelley TG, Parthasarathy Rao P and Walker TS. 1993. TheRelative Value of Cereal Straw Fodder in India: Implicationsfor Cereal Breeding Programs at ICRISAT, pp. 88-105 in

Dorward A, Poole N, Morrison J, Kydd J and Urey I. 2003.Markets, Institutions and Technology: Missing Links inLivelihoods Analysis. Development Policy Review, 2003,21(3): 319–332.

Ellis F. 1999. Rural Livelihood Diversity In DevelopingCountries: Evidence And Policy Implications NaturalResource Perspectives Number 40, April 1999. London,UK: Overseas Development Institute.

Ellis F. 2000. Rural livelihoods and diversity in developingcountries. Oxford: Oxford University Press.

Fan S and Hazell P. 2000. Should India invest more in Less-Favored Areas. Environment and Production TechnologyDivision (EPTD) Discussion Paper.25. Washington DC.International Food Policy Research Institute

Fan S, Hazell P and Thorat S. 1998. GovernmentSpending, Growth and Poverty: An Analysis of Interlinkagesin Rural India. Environment and Production TechnologyDivision (EPTD) Discussion Paper no. 33. Washington DC.International Food Policy Research Institute

Fan S, Hazell P and Haque T. 2000. Targeting publicinvestments by agro-climatic zone to achieve growth andpoverty alleviation goals in rural India. Food Policy, Vol. 25,No, 4, pp. 411-427.

Farrington J. 2001. Sustainable Livelihoods, Rights and TheNew Architecture of Aid. Natural Resource Perspectives.No 69. London: Overseas Development Institute.

Foster J. 1984. On economic poverty: ‘A survey of aggregatemeasures’, Advances in Econometrics, 3:215-251

Francis P, Milimo JT, Njobvu CA and Tembo SPM. 1997.Listening to Farmers: Participatory Assessment of PolicyReform in Zambia’s Agriculture Sector. World BankTechnical Paper 375, Washington D.C.: World Bank.

Gaiha R. 1989. ‘Are the chronically poor also the poorest inIndia’, Development and Change 20(2), 295–322.

Gaiha R and Imai K. 2003. Vulnerability, shocks andpersistence of poverty – Estimates for Semi-Arid ruralSouth India.

GOI (Government of India). 1973. Task Force Report1973, Planning Commission, New Delhi.

GOI (Government of India). 2002. Annual Report 2001-2002, Ministry of Rural Development, New Delhi, 2002.

Gujarati DN. 1995. Basic Econometrics, 3/e. Singapore:McGraw-Hill. 838 pp.

Gulati A and Tim Kelley. 1999. Trade Liberalization andIndian Agriculture. New Delhi, India: Oxford UniversityPress. Pp 15–36.

Hardin G. 1968. The Tragedy of the Commons. Science162:1243–48.

Hewitt de Alcantara C (ed.). 1993. Real markets: Socialand political issues of food policy reform United NationsResearch Instituted for Social Development

Bibliography

52

Overcoming Poverty in Rural India

Scoones I and Thompson J (eds.). 1994. Beyond FarmerFirst: Rural People’s Knowledge, Agricultural Research andExtension Practice. London: Intermediate TechnologyPublications.

Sen AK. 1981. Poverty and Famines: An Essay onEntitlement and Destitution. Oxford: oxford UniversityPress.

Sen AK. 1984. Commodities and capabilities, NorthHolland, Amsterdam, The Netherlands.

Sen AK. 1990. Gender and cooperative conflicts. inI.Tinker (ed.) Persistent inequalities: women and worlddevelopment, Oxford University Press, U.K.

Shah T. 1993. Groundwater Markets and IrrigationDevelopment: Political Economy and Practical Policy,Oxford University Press, Delhi.

Sharma M, Brown L, Qureshi A and Garcia M. 1995. AnEcoregional Perspective on Malnutrition, IFPRI 2020 VisionBrief, 14, April 1995

Singh K. 1988. Managing Dry Land WatershedDevelopment Programmes; Lessons of KarnatakaExperience, Research Paper 1, Institute of RuralManagement, Anand.

Singh K. 1991a. Dryland Watershed Development andManagement: A Case Study in Karnataka, Indian Journal ofAgricultural Economics, Vol. 46, No. 2, April-June, pp. 121-131.

Singh K. 1991b. Determinants of People’s Participation inWatershed Development and Management: An ExploratoryCase Study. Indian of Agricultural Economics, Vol. 46, No.3, July-September, pp. 278-286.

Singh K. 1994a. People’s Participation in Micro WatershedManagement – A Case Study of an NGO in Gujarat, IndianJournal of Soil Conservation, Vol. 22, No. 1-2, pp. 271-78.

Singh K. 1994b. Managing Common Pool NaturalResources: Principles and Case Studies, Oxford UniversityPress, New Delhi.

Singh K. 1995a. The Watershed Approach to Sustainabilityof Renewable Common Pool Natural Resources: Lessons ofIndia’s Experience, Research Paper 14, Institute of RuralManagement, Anand.

Singh K. 1995b. Co-operative Property Rights as anInstrument of Managing Groundwater. in Moench (ed.)Groundwater Law: The Growing Debate, VIKSAT – PacificInstitute Collaborative Groundwater Project, Ahmedabad,Gujarat, India.

Singh K. 1995c. People’s Participation in ManagingCommon Pool Natural Resources. Social Change, Vol. 25,No. 1, March.

Singh K. 1997a. Measurement of Poverty: A Critique ofCommon Methods and Indicators. Book chapter in R.C.

Social Science Research for Agricultural TechnologyDevelopment: Spatial and Temporal Dimensions, Dvorak,K. (ed.). CABI, London.

Kerr J. 2000. Development Strategies For Semiarid SouthAsia 2020 Focus 4 (Promoting sustainable development inless-favored areas), Brief 6 of 9, November, Washington:IFPRI

Laxmiah A and Vijayaraghavan K. 2003. Health Action,April, pp 13-15.

Lockwood M. 1995. Beyond the feminisation of poverty:gender-aware poverty Reduction. BRIDGE Developmentand Gender in Brief, 2, pp.1-3.

Marshland N and Parthasarathy Rao P. 1999. Marketing ofkharif and rabi Sorghum in Andhra Pradesh, Karnataka, andMaharashtra. Socioeconomics and Policy Program, ProgressReport. ICRISAT, Patancheru. (Being finalized).

Mehta AK and Shah A. 2003. Chronic poverty in India:Incidence, causes and policies. World Development31(3):491–511.

Mortimore M, Adams B and Harris F. 2000. Poverty AndSystems Research In The Drylands. Gatekeeper Series no.94, London: International Institute for Environment andDevelopment

Padmaja R. 1998. Women in SAT Agriculture.Socioeconomics and Policy Program Working Paper,International Crops Research Institute for the Semi-AridTropics.

Parthasarathy Rao P and Birthal PS. 2002. Crop-livestocksystems research and policy issues: India. Policy Brief 1.Strategic assessment and development pathways foragriculture in the Semi-Arid Tropics. ICRISAT

Parthasarathy Rao P, Birthal PS, Joshi PK and Kar D.2002. Agricultural diversification in India, role ofurbanization and factors influencing diversification. Paperpresented at the workshop on agricultural diversification inSouth Asia, Jointly organized by RGOB, MoA, ICAR andIFPRI, Paro, Bhutan, 21-23 November 2002.

Poulton C, Al-Hassan R, Cadisch G, Reddy C and SmithL. 2001. The cash crop versus food crop debate. IssuesPaper – 3, January 2001 Crop Post-Harvest Programme,Department for International Development (DFID).

Putnam RD. 1993. Making Democracy Work: CivicTraditions in Modern Italy. Chichester: PrincetonUniversity Press.

Ryan JG and Spencer DC. 2001. Future challenges andopportunities for Agricultural R &D in Semi-Arid Tropics,International Crops Research Institute for the Semi-AridTropics

Scoones I. 1998. Sustainable Rural Livelihoods: AFramework for Analysis. IDS Working Paper No. 72.Brighton: Institute of Development Studies at theUniversity of Sussex, June.

Choudhury and S. Rajakutty (Eds.), Fifty Year of RuralDevelopment in India: Retrospect and Prospect, Vol. 1,National Institute of Rural Development, Hyderabad, pp.183-196.

Singh K. 1997b. Property Rights and Tenures in NaturalResource, Book chapter in John Kerr, Dinesh Marothia,Katar Singh, C. Ramasamy and William Bentley (eds.)Natural Resource Economics: Theory and Applications inIndia, Oxford & IBH Publishing Co. New Delhi, 1997.

Singh K. 1999a. Sustainable Development: SomeReflections. Indian Journal of Agricultural Economics, 54 (1).

Singh K. 1999b. Rural Development: Principles, Policiesand Management, Second edition, Sage Publication, NewDelhi.

Singh K. 2001a. Managing India’s Water Resources: SomePolicy Issues and Options. Paper presented at the CurtainRaiser Function of Water Resources Day 2001 and WorldWater Day organized by the Indian Water Resources Society(Delhi Centre) on March 22, 2001 in Teen MurtiAuditorium, New Delhi.

Singh K. 2001b. Managing India’s Water Resources: SomePolicy and Institutional Issues and Strategies for ResolvingThem. Proceedings of Curtain Raiser Function of WaterResources Day, 2001 and World Water Day, Indian WaterResources Society (Delhi Centre), New Delhi.

Singh K and Ghatak RN. 1995. Pricing of Canal IrrigationWater: An Application of the Contingent Valuation Method,Arhta Vikas, July–December.

Singh K and Gupta KK. 1998. The Sadguru Model ofCommunity-Based Natural Resources Management.Occasional Publication No.14, Institute of RuralManagement, Anand.

Start D, Deshingkar P and Farrington J. (forthcoming).The influence of wealth on patterns of livelihooddiversification – evidence from field studies in AndhraPradesh and Madhya Pradesh to be published in a book onthe rural non-farm economy published by IHD, New Delhi.

Swift J. 1989. Why are rural people vulnerable to famine?IDS Bulletin, 20, 2.

Thirtle C, Beyers L, Lin L, McKenzie-Hill V, Irz X,Wiggins S and Piesse J. 2002. The impact in changes inagricultural productivity on the incidence of poverty indeveloping countries. Report to Department forInternational Development (DFID) No 7946.

Thomas D, Zerbini E, Parthasarathy Rao P andVaidyanathan A. 2002. Increasing animal productivity onsmall mixed farms in South Asia: A system perspective.Agricultural Systems 71 (2002) 41-57

UNDP (United Nations Development Programme). 1997.Sustainable livelihoods: Concepts, principles and approachesto indicator development. A draft discussion paper. NewYork: United Nations Development Programme.

Uwe Deichmann. 1999. Geographic aspects of inequalityand poverty, January 1999, http://www.worldbank.org/poverty/inequal/povmap/povmap.pdf

Wood GD. 1992. Parallel rationalities in service provision:the general case of corruption in rural development. Paperto European Network of Bangladesh Studies conference,CDR, Copenhagen.

World Bank. 1999. Can Anyone Hear Us? Voices from 47countries. Poverty Group, PREM, Washington DC, USA:World Bank.

53

Bibliography

Appendixes

Appendix 1. The Semi-Arid Tropics and OtherAgro-Ecological Zones

Appendix 1.1. The semi-aridtropicsThe Technical Advisory Committee (TAC) of theConsultative Group on International AgriculturalResearch (CGIAR) and FAO defines SAT as thoseareas, which have (a) (crop) growing period of 75–180days; (b) mean monthly temperature higher than 18degrees Celsius for all the twelve month of the year;and (c) daily mean temperature during the growingperiod higher than 20 degrees Celsius (Ryan andSpenser 2001). According to this definition, India hasa SAT area of 1,289,713 sq km, which was nearly 42%of the total geographical area of the country (Ryan andSpenser 2001). Troll (1964) as quoted in Gulati and

Kelley (1999) defines SAT as those tropical regionswhere rainfall exceeds potential evapotranspiration fortwo to seven months a year. Gulati and Kelley (1999)have used a slightly modified version of the agro-ecological region (AER) classification system used bythe National Bureau of Soil Survey and Land UsePlanning (NBSS&LUP) of the Indian Council ofAgricultural Research (ICAR). The NBSS&LUP hasdivided India into 20 AERs, using soil and various bio-climatic parameters. Following AER classification,Gulati and Kelley delineated SAT by superimposingthe AER map on to the All-India district map.Districts, which had 50% or more of their land areafalling within the AERs 2 through 10, were consideredas constituting SAT.

57

58

Overcoming Poverty in Rural India

Appendix 1.2. List of districts by AEZ and irrigation levels

HIGH Irrigation MEDIUM Irrigation LOW Irrigation(>40% of GCA) (25–40% of GCA) (<25% of GCA)

AEZ State/District Name State/District Name State/District Name

Humid(LGP:>=180 days)

Andhra PradeshSrikakulam, East Godavari,West Godavari, Khammam

Andhra PradeshVijayanagaram,Visakhapatnam

AssamDhubri, Kokrajhar,Bongaigaon, Goalpara,Golaghat, Nalbari,Kamrup, Darrang,Sonitpur, Lakhimpur,Dhemaji, Morigaon,Nagaon, Barpeta,Jorhat, Sibsagar,Dibrugarh, Tinsukia,Karbi Anglong, NorthCachar Hills,Karimganj, Hailakandi,Cachar

BiharGopalgunj, Champaran(West), Champaran (East),Saharsa, Madhepura, Purnea,Katihar, Khagaria, Supaul

BiharSeethamarhi, Darbhanga,Madhubani, Bhagalpur,Araria, Banka

BiharGodda, Sahebganj,Dumka, Dhanbad,Hazaribagh,Lohardagga, Gumla,Ranchi, Singhbhum(West), Singhbhum(East), Kishangunj,Bokaro, Chatra

Himachal PradeshKinnaur

Himachal PradeshKangra, Solan, Sirmaur

Himachal PradeshChamba, Hamirpur,Una, Bilaspur, Mandi,Kullu, Simla

KarnatakaShimoga

KarnatakaDakshina Kannara

KarnatakaChickmagalur, Hassan,Kodugu, UttaraKannara

KeralaThrissur

KeralaKasaragod, Ernakulam,Alappuzha/Allepey

KeralaKannur, Wayanad,Kozhikode,Malappuram, Palakkad,Idukki, Kottayam,Pathanamthitta,Kollam/Quilon,Thiruvananthapuram

Madhya PradeshNarsimhapur, Balaghat

Madhya PradeshMandla, Seoni,Rajnandgaon, Bastar

MaharashtraBhandara

MaharashtraGadchiroli

MaharashtraThane, Raigad,Ratnagiri, Sindhudurg,Kolhapur, Chandrapur

59

HIGH Irrigation MEDIUM Irrigation LOW Irrigation(>40% of GCA) (25–40% of GCA) (<25% of GCA)

AEZ State/District Name State/District Name State/District Name

MaharashtraThane, Raigad, Ratnagiri,Sindhudurg, Kolhapur,Chandrapur

OrissaCuttack, Balangir, Ganjam,Puri, Bhadrah, Jagatsinghpur,Jajpur, Kendrapara, Nayagarh,Khurda, Gajapati, Sonepur

OrissaKendujhar,Mayurbhanj,Dhenkanal, Phulbani,Kalahandi, Koraput,Boudh, Nawapara,Malkangiri,Nawarangapur,Rayagada, Angul

Tamil NaduChengalpattu MGR,Cuddalore (S. Arcot),Villupuram RamaswamyPadaiyatchiar

Tamil NaduThe Nilgiris

Uttar PradeshDehradun, Nainital,Haridwar, Kheri,Mahrajgunj, Gorakhpur,Deoria

Uttar PradeshBahraich, Gonda,Sidharthnagar

Uttar PradeshTehri Garhwal,Garhwal, Pithorgarh,Almorah

West BengalHooghly, Bankura,Burdwan, Birbhum

West BengalMidnapore, Purulia

West BengalJalpaiguri,Darjeeling, WestDinajpur, Malda,Murshidabad, Nadia,24-Paraganas (N),24-Paraganas (S),Howrah

Semi Arid Temperate(LGP:75-179 days,Temp:<=18o C)

BiharSaran, Sivan, Begusarai,Samastipur

BiharMuzaffarpur, Vaisali

HaryanaAmbala,Yamunanagar,Kurukshetra, Kaithal,Karnal, Panipat, Sonipat,Faridabad, Gurgaon,Rewari, Mahendragarh, Jind

Madhya PradeshMorena, Gwalior, Datia,Tikamgarh

Madhya PradeshBhind, Chhatarpur

Madhya PradeshPanna, Satna, Rewa

PunjabGurdaspur, Amritsar,Ludhiana, Jalandhar,Kapurthala, Hoshiarpur,Roopnagar, Patiala

RajasthanJaipur, Dausa

RajasthanAlwar, Bharatpur, Dholpur,Sawai Madhopur, Sikar, Baran

RajasthanAlwar, Bharatpur,Dholpur, SawaiMadhopur, Sikar,Baran

Appendixes

60

Overcoming Poverty in Rural India

HIGH Irrigation MEDIUM Irrigation LOW Irrigation(>40% of GCA) (25–40% of GCA) (<25% of GCA)

AEZ State/District Name State/District Name State/District Name

Uttar PradeshBijnor, Moradabad,Rampur, Saharanpur,Muzaffarnagar, Meerut,Ghaziabad, Bulandshahar,Aligarh, Mathura, Agra,Ferozabad, Etah, Mainpuri,Budaun, Bareilly, Pilibhit,Shajahanpur, Sitapur,Hardoi, Unnao, Lucknow,Rae Bareily, Farukkabad,Etawah, Kanpur (Dehat),Kanpur (Nagar), Jhansi,Fatehpur, Pratapgarh,Allahabad, Barabanki,Faizabad, Sultanpur, Basti,Azamgarh, Jaunpur, Ballia,Gazipur, Varanasi,Mirzapur

Uttar PradeshJalaun, Hamirpur, Banda

Uttar PradeshJalaun, Hamirpur,Banda

Semi-Arid Tropic (LGP:75-179 days, Temp>18o C)

Andhra PradeshKrishna, Guntur, Nellore,Chittoor, Cuddapah,Nizamabad, Karimnagar,Warangal, Nalgonda

Andhra PradeshPrakasam, Mahabubnagar,Ranga Reddy, Medak

Andhra PradeshKurnool, Adilabad

BiharPatna, Nalanda, Bhojpur,Rohtas, Aurangabad,Jahanabad, Gaya, Nevada,Mungair, Buxar, Babhua,Jamui

BiharPalamau, Gadva

BiharDevghar, Giridih,Giridih

GujaratMehsana, Gandhinagar,Kaira, Sura

GujaratBhavnagar, Banaskantha,Sabarkantha, Ahmedabad,Vadodara, Valsad

GujaratJamnagar, Rajkot,Surendranagar,Amreli, Junagadh,Panch Mahals,Bharuch

KarnatakaMandya

KarnatakaBelgaum, Chitradurga,Mysore, Raichur

KarnatakaBangalore (Urban),Bangalore (Rural),Bidar, Dharwad,Gulbarga, Kolar,Tumkur

Madhya PradeshHoshangabad

Madhya PradeshShivpuri, Mandsaur, Ratlam,Ujjain, Shajapur, Dewas,Dhar, Indore, West Nimar,East Nimar, Rajgarh,Sehore, Bilaspur (MP),Durg, Raipur

Madhya PradeshGuna, Sagar, Damoh,Shahdol, Sidhi, Jhabua,Vidisha, Raisen, Betul,Jabalpur, Chhindware,Surguja, Raigarh

61

HIGH Irrigation MEDIUM Irrigation LOW Irrigation(>40% of GCA) (25–40% of GCA) (<25% of GCA)

AEZ State/District Name State/District Name State/District Name

RajasthanSirohi, Bundi, Kota

RajasthanAjmer, Pali, Bhilwara,Udaipur, Chittorgarh,Banswara, Jhalawar,Rajsamand

MaharashtraSatara

MaharashtraNasik, Dhule, Jalgaon,Ahmadnagar, Pune,Sangli, Solapur,Aurangabad, Jalna,Parbhani, Beed,Nanded, Osmanabad,Latur, Buldhana,Akola, Amravati,Yeotmal, Wardha,Nagpur

Uttar PradeshLalitpur

OrissaSundargarh

Tamil NaduVellore (N. Arcot),Tiruvannamalai, Erode(Periyar), Coimbatore,Trichy, Tanjavur,Puddukottai, Madurai,Tirunelveli Kattabom,Kanyakumari, NagaiQuaid-E-Milleth

Tamil NaduSalem, Dindigul,Ramanthapuram,Kamarajar

Arid (LGP: 0-74 days) HaryanaRohtak, Bhiwani, Hissar,Sirsa

KarnatakaBellary

Andhra PradeshAnanthapur

PunjabFerozpur, Sangrur,Bhatinda, Faridkot, Mansa

GujaratKutch

RajasthanGanganagar

RajasthanJalore

RajasthanBikaner, Churu,Jhunjhunu, Jaisalmer,Jodhpur, Nagaur,Barmer

Uttar PradeshChamoli, Uttar Kashi

Appendixes

62

Overcoming Poverty in Rural India

Appendix 1.3. Relative importance of SAT with respect to thevarious climatic zones of India

Table 1.3.1. Relative importance of SAT and non-SAT districts of India: 1997–98.

Share of All India (%)

Item SAT Non-SAT

Geographical area 37.2 62.8Population 36.9 63.1Net cropped area (NCA) 46.2 53.8Gross cropped area (GCA) 42.9 57.1Gross irrigated area (GIA) 31.9 68.1Coarse cereals (area) 58.7 41.3Pulses (area) 52.6 47.4Oilseeds (area) 59.7 40.3Commercial crops (area) 60.0 40.0Fruits and vegetables (area) 28.6 71.4Coarse grains (production) 60.5 39.5Pulses (production) 51.5 48.5Oilseeds (production) 62.8 37.2Value of Agricultural production (crops + livestock) 36.9 63.1

Source: ICRISAT database

Table 1.3.2. Relative importance of various climatic zones of India: 1970–71.

Share to total (%)*

Importance/regions Arid Humid Semi-arid temperate Semi-arid tropics

Geographical area 12.3 30.0 13.8 43.9Population 4.0 33.6 21.9 40.5Net cultivated area 10.9 21.4 19.4 48.3Gross cropped area 10.3 22.8 21.6 45.4Gross irrigated area 11.2 19.3 35.7 33.9Coarse cereals area 14.9 8.2 19.3 57.6Pulses area 12.3 13.1 22.0 52.6Oilseeds area 7.7 12.4 13.6 66.2Commercial crops area 10.8 5.8 14.8 68.6Fruits and vegetables area 5.0 26.2 18.6 50.2Production of coarse grains 7.3 27.4 24.2 41.1Production of pulses 4.2 30.1 35.0 30.7Production of oilseeds 13.8 19.5 11.8 54.9VOP agriculture 6.9 29.9 24.2 39.1

*Refers to the sum of 16 major agricultural states of India

Source: ICRISAT database

63

Table 1.3.3. Selected indicators of SAT and non-SAT districts of India 1997–98.

Semi-arid Semi-aridImportance/regions Arid Humid temperate tropics All India

Population density 132 405 586 300 324(No./Sq. km of geog. Area)

Literate rural female (%) 36 41 35 39 39

Urban population% 24 20 24 30 28

Land size (ha) 4.5 1.0 1.1 1.9 1.6

Livestock units per ha (NCA) 0.9 2.8 2.1 1.7 2

Livestock Units Per capita 0.39 0.28 0.28 0.33 0.31

No. of small land holders 37 87 83 69 78

Net Area Sown Per Capita 0.038 0.010 0.012 0.018 0.015per 100 person

Per capita gross value of agricultural 7480 3215 4606 3919 3989produce (Rs person-1)

Gross value of agricultural produce 19613 31570 39310 21985 27209(Rs ha-1)

Tractor density (No/10’000 ha) 9.3 3.9 20.6 4.5 8.3

Diesel & Electric pump sets density 26.8 58.5 125.9 74.2 74.9(No/10’000 ha)

Cropping intensity (%) 126.6 140.3 148.0 124.2 139.2

HYV crop area (%) 26.5 41.2 51.5 31.7 40.7

Gross irrigated (%) 42.5 30.4 65.4 28.9 38.3

Fertilizer (kg ha-1 of NCA) 78 104 168 97 101

Fertilizer (kg ha-1 of GCA) 62 74 113 78 85

Mean normal rainfall (mm) 483 1835 956 965 1212

Market density (No/10,000 sq km) 14.1 23.8 34.2 21.3 22.1

Road density (km/sq. km) 0.3 0.7 0.5 0.5 0.75

Source: ICRISAT Database

Appendixes

64

Overcoming Poverty in Rural India

Table 1.3.4. Selected indicators of SAT and Non-SAT districts of India 1970–71.

Indicators SAT Non-SAT Total

Population density (No./sq km of geog. Area) 160 206 185Literate rural female (%) 12.0 13.6 13.0Urban population% 20.7 15.9 17.9Land size (ha) 3.7 1.9 2.8Livestock units per ha (NCA) 1.5 1.9 1.7No. of small land holders 50.2 74.2 62.0Tractor density (No/10,000 ha) 0.5 1.7 1.1Diesel & Electric pump sets density (No/10,000 ha) 31.4 17.7 24.2Cropping intensity (%) 111.1 124.8 118.2HYV crop area (%) 8.7 9.9 9.4Gross irrigated (%) 17.8 30.0 24.4Fertilizer (kg ha-1 of NCA) 16.9 21.0 19.0Fertilizer (kg ha-1 of GCA) 15.2 16.8 16.1Market density (No/10,000 sq km) 13.7 15.7 14.7Road density (km/sq. km) 0.3 0.2 0.2

Source: ICRISAT Database

Appendix 2. Evolution and Current Thinkingof the Sustainable Livelihoods Approach

2.1 The genesis, evolution andcurrent thinkingThe Sustainable Livelihoods Approach (SLA) is nowused by a number of governments and internationaldevelopment agencies including the Department forInternational Development (DFID), United NationsDevelopment Programme (UNDP), SIDA, CAREand the International Fund for AgriculturalDevelopment (IFAD) as their overarching frameworkfor poverty reduction. The concept of sustainablelivelihoods was arguably first introduced by theBrundtland Commission on Environment andDevelopment, and then advocated by the 1992 UNConference on Environment and Development as abroad goal for poverty eradication. It was around thattime that Chambers and Conway (1992) defined alivelihood and this is still the underlying premise of theSLAs today:

A livelihood comprises the capabilities, assets(stores, resources, claims and access) andactivities required for a means of living: alivelihood is sustainable which can cope with andrecover from stress and shocks, maintain orenhance its capabilities and assets, and providesustainable livelihood opportunities for the nextgeneration; and which contributes net benefits toother livelihoods at the local and global levels andin the short and long term.

The sustainable livelihoods framework (as opposedto the approach) is an analytical tool, which provides astructure to help us understand livelihoods and ensurethat external support is congruent with peoples’livelihood strategies and priorities.

2.2 Applying the SL approach toSAT agriculture – understandingrisk and vulnerabilityUnderstanding complexity: SAT livelihoods arecomplex and differ from place to place and evenbetween individuals and households. There are bigdifferences between the poor and the better-off inrelation to the sources of income that feature moststrongly in their respective livelihood strategies (Ellis

1999) The focus on assets also alerts us to thepossibility that policies and projects targetingindividuals or households with more assets, are likelyto improve the incomes of those who are alreadybetter-off.

SAT livelihoods are typically a changeablecombination of farm work, non-farm work andmigrant remittances. The importance of migration(both temporary and permanent) needs to berecognized and has far-reaching implications forpoverty in SAT areas. There is growing evidence thatmigration may be accumulative and remittances are animportant part of the economy in remote, dry, ruralareas (Deshingkar and Start 2003).

Understanding Change: It is important to understandthat livelihoods progress in a non-linear, iterativefashion. The SLA offers scope for a ‘dynamic’approach in that it attempts to understand change overtime. It calls for a process approach in research andintervention implementation, which means updatingexisting knowledge all the time and generating newknowledge about livelihoods.

Taking a multi-dimensional view of poverty: The SLArecognizes the need for a more ‘holistic’ and ‘non-sectoral’ approach, which is multidisciplinary withmultiple actors and influences because it is dealingwith complex livelihoods that cannot be understoodthrough one perspective. It is essentially a ‘people-centred’ approach, and puts the perceptions andpriorities of the poor first because outside expertsusually have very different perceptions from thosewho actually experience poverty This has fundamentalimplications for the understanding and measurementof poverty. Poverty should not be viewed only in termsof income or quantifiable indicators but also in termsof other less tangible factors such as status, happiness,safety, dignity and autonomy. In fact recent definitionsof poverty include qualitative dimensions because it isnow widely recognized that it is important thatintangible processes and resources are incorporatedinto assessments of poverty (see for exampleparticipatory poverty assessments conducted by theWorld Bank year).

Adato and Dick (2002) for example argue thatconventional poverty measures based on income,consumption, or nutrition are broadened to address

65

66

Overcoming Poverty in Rural India

additional aspects of poverty and well-being, eg, accessto land, water, credit, or education, vulnerability tonatural disasters, political rights, physical safety, andsocial relationships that provide economic security andsocial well-being. Adato and Meinzen-Dick identifythe following aspects of SL as being relevant tointegrating social and economic assessment in the areaof agricultural research and technology development:• Defining poverty more broadly than by simple

income or consumption standards, to include otheraspects of well-being and empowerment;

• Distinguishing the impacts of agricultural researchon socially differentiated groups;

• Attention to the importance of vulnerability and awhole range of assets in influencing adoption andimpact of new technologies;

• Consideration of the nature of research anddissemination processes and how they impact onpoverty; and

• Recognition that households and individuals are notonly ‘farmers’ or ‘consumers’, but may employ arange of livelihood strategies, and that thisinfluences the impact of agricultural research.The following checklist, which builds on the one

proposed by Scoones (1998) is useful for diagnosingthe nature, extent and severity of poverty:• Who are the poor – by caste, gender, and class – An

important point here is to bear in mind that thepoor are not just farmers. Much of the literature ondryland farming systems and SAT agriculturerelates to the management of constraints andopportunities by smallholders. In the Indiancontext, an approach to SAT would not becomplete without addressing the specificconstraints, priorities and livelihood strategies ofassetless laborers who form close to 40% of theworking population in several locations. In fact thecontinued supposition that the poor are mainlypoor farmers is inaccurate and even dangerousbecause it excludes a majority of those who need tobe helped out of poverty.

• What are the main livelihood strategies pursued bydifferent groups of people? An important guide foragriculture research is that a priori assumptionsabout the survival strategies of rural householdscannot be made. It has been argued that assuming aparticular rural social group is mainly dependent onthe production of a particular crop or farmingsystem for survival is likely to be wrong. In realitylivelihood strategies are a great deal morecomplicated than assumed (Ellis 1999).

• Sequencing – What is the starting point for

successfully establishing a particular livelihoodstrategy? Is one type of livelihood resource anessential precursor for gaining access to others?

• Substitution – Can one type of capital besubstituted for others? Or are different forms ofcapital needed in combination for the pursuit ofparticular livelihood strategies?

• Clustering – If you have access to one type ofcapital, do you usually have access to others? Or isthere a clustering of particular combinations oflivelihood resources associated with particulargroups of people or particular livelihood strategies?

• Access – Different people clearly have differentaccess to different livelihood resources. A sociallydifferentiated view to analyzing livelihoods istherefore critical, one that disaggregates the chosenunit of analysis – whether community, village orhousehold – and looks at individuals or groups ofsocial actors and their relationships, in relation tothe range of relevant dimensions of difference(wealth, gender, age and so on) and the distributionof control over resources.

• Trade-offs – In pursuing a particular portfolio oflivelihood strategies, what are the trade-offs facedby different people with different access todifferent types of livelihood resource? Dependingon who you are, differential access to differenttypes of capital may have positive or negativeimplications in terms of the success or otherwise inthe pursuit of a sustainable livelihood.

• Trends – What are the trends in terms of availabilityof different types of livelihood resource? How aredifferent capital assets being depleted andaccumulated, and by whom? What are the trends interms of access? What new livelihood resources arebeing created through environmental, economicand social change?

• Power – who holds power over investmentdecisions, resource allocation, providing access toservices?

• Voice – How can we empower the poor and givethem more voice to demand their entitlements andchallenge power structures?

• Technology – which needs of the poor (expressed intheir own terms) are amenable to technicalsolutions and do they have the capacity to utilizetechnologies that may be developed?

• Markets – which markets are important for thepoor and how can interventions help to createmarkets or fill gaps where markets are missing? Forinstance are interlocked markets pro-poor in somecircumstances?

Table 3.2. Water availability and infrastructure indices.

Rainfall Irrigation Fallow Market Road InfrastructureAgro-ecological zone (mm) ratio land (%) density density index

Humid high 1003 56.0 9.88 36.1 0.71 99.9Humid medium 1273 34.3 9.26 20.8 0.99 81.7Humid low 1768 16.3 15.09 23.3 0.63 83.6Humid 1495 31.0 12.46 25.7 0.72 87.0Semi-Arid Temperate high 776 71.7 9.13 29.9 0.18 104.2Semi-Arid Temperate medium & low 727 33.6 9.32 18.9 0.11 82.4SEMI-ARID TEMPERATE 766 62.1 9.17 26.9 0.16 98.3Semi-Arid Tropic high 815 56.8 24.54 29.1 0.81 105.8Semi-Arid Tropic medium 1009 31.1 16.64 21.3 0.35 86.9Semi-Arid Tropic low 924 15.6 11.64 21.7 0.43 86.0Semi-Arid Tropic 919 28.4 15.98 23.3 0.49 90.9Arid high 393 79.3 5.61 58.6 0.39 128.9Arid low & medium 587 20.5 20.17 12.5 0.17 85.3Arid 506 39.9 15.88 22.3 0.22 94.6All India 1099 36.0 14.0 24.4 0.49 91.0

Market density: Number per 10,000 per sq kmRoad density: Per km2

Source: ICRISAT Database, 1998

Table 3.1. Distribution of population across agro-ecological zones.

Rural UrbanPopulation Total

Number % to all Number % to all Number % to allAgro-ecological zone (millions) India (millions) India (millions) India

Humid High 70.387 10.2 14.452 6.2 84.839 9.2Humid Medium 57.010 8.2 10.728 4.6 67.738 7.3Humid Low 123.077 17.8 33.364 14.4 156.441 16.9Humid 250.474 36.2 58.544 25.2 309.018 33.4Semi-Arid Temperate High 122.133 17.7 39.611 17.0 161.744 17.5Semi-Arid Temperate medium & low 23.599 3.4 3.637 1.6 27.236 2.9Semi-Arid Temperate 145.732 21.1 43.248 18.6 188.980 20.4Semi-Arid Tropic High 81.316 11.8 34.089 14.7 115.405 12.5Semi-Arid Tropic Medium 75.845 11.0 28.343 12.2 104.188 11.3Semi-Arid Tropic Low 90.315 13.1 0.492 0.2 90.807 9.8Semi-Arid Tropic 247.476 35.8 62.924 27.1 310.400 33.6Arid High 1.161 0.2 2.980 1.3 4.141 0.4Arid Low & Medium 13.979 2.0 1.012 0.4 14.991 1.6Arid 15.140 2.2 3.993 1.7 19.132 2.1Unclassified1 32.692 4.7 63.688 27.4 96.651 10.5All India 691.784 100.0 232.397 100.0 924.181 100.0

1. Information for classification was not available for some districts and some Urban districts were not included.Source: ICRISAT Database, 1998.

Appendix 3. Determinants of Poverty

67

68

Overcoming Poverty in Rural India

Table 3.3. Incidence of rural poverty across agro-ecological zones.

SquaredAgro-ecological zone Head count Poverty gap poverty gap Sen’s Index MPCE

Humid High 22.1 3.7 7.1 5.2 462.0Humid Medium 21.6 4.0 8.7 5.5 458.6Humid Low 25.6 5.0 9.9 6.9 487.2Humid 23.7 4.4 8.8 6.1 473.6Semi-Arid Temperate High 13.7 2.1 3.8 3.0 512.1Semi-Arid Temperate medium & low 19.5 2.8 4.1 3.8 462.8Semi-Arid Temperate 14.6 2.2 3.8 3.1 504.1Semi-Arid Tropic High 24.0 4.4 8.1 6.1 480.7Semi-Arid Tropic Medium 23.2 4.1 8.5 5.7 468.5Semi-Arid Tropic Low 25.5 4.7 9.2 6.6 467.8Semi-Arid Tropic 24.3 4.4 8.6 6.2 472.3Arid High 2.8 0.3 0.5 0.5 738.9Arid Low & Medium 13.4 2.1 4.0 3.0 532.2Arid 12.6 2.0 3.7 2.8 548.1All India 21.3 3.8 7.4 5.3 485.9

Note: MPCE: Monthly Per Capita Total Expenditure in rupees.Source: ICRISAT Database, 1998 and National Sample Survey 1999–2000

Table 3.4. Distribution of rural poor across agro-ecological zones.

Poor on head count basis

Agro-ecological zone Number (millions) Percentage to total poor

Humid high 15.591 10.6Humid medium 12.304 8.3Humid low 31.479 21.3Humid 59.374 40.3Semi-Arid Temperate high 16.783 11.4Semi-Arid Temperate medium & low 4.604 3.1Semi-Arid Temperate 21.387 14.5Semi-Arid Tropic high 19.478 13.2Semi-Arid Tropic medium 17.632 12.0Semi-Arid Tropic low 23.071 15.6Semi-Arid Tropic 60.180 40.8Arid high 0.309 0.2Arid low & medium 1.874 1.3Arid 2.182 1.5All India 147.478 100.0

Source: National Sample Survey 1999-2000

69

Table 3.5. Distribution of poor according to severity of poverty (percentages).

Agro-ecological zone Very poor Moderately Poor Non poor Rich

Humid high 5.3 16.9 39.7 38.1Humid medium 6.3 15.3 41.5 36.9Humid low 7.9 17.7 38.0 36.4Humid 6.8 16.9 39.3 37.0Semi-Arid Temperate high 2.6 11.1 35.2 51.1Semi-Arid Temperate medium & low 2.8 16.7 38.3 42.2Semi-Arid Temperate 2.6 12.0 35.7 49.7Semi-Arid Tropic high 6.7 17.2 39.1 37.0Semi-Arid Tropic medium 5.8 17.4 40.4 36.4Semi-Arid Tropic low 7.2 18.3 37.9 36.5Semi-Arid Tropic 6.6 17.7 39.1 36.6Arid high 0.3 2.4 20.8 76.4Arid low & medium 2.2 11.2 36.2 50.4Arid 2.1 10.5 35.0 52.4All India 5.6 15.7 38.0 40.7

Note: Very Poor: MPCE less than 75% of Poverty LineModerately Poor: MPCE between 75% and 100% of Poverty LineNon Poor: MPCE between 100% and 150% of Poverty LinRich: MPCE above 150% of Poverty LineSource: National Sample Survey 1999-2000

Table 3.6. Incidence of head count poverty across social groups.

Scheduled tribes Scheduled castes Others

Incidence Population Incidence Population Incidence PopulationAgro-ecological zone (%) (%) (%) (%) (%) (%)

Humid high 37.06 5.5 30.56 13.1 17.88 11.1Humid medium 39.88 5.7 30.21 8.2 17.52 9.4Humid low 48.30 35.4 26.61 12.3 19.54 20.8Humid 43.2 44.3 28.5 33.3 18.6 41.3

Semi-Arid Temperate high 11.88 0.6 22.30 16.9 10.88 12.5Semi-Arid Temperate 15.78 1.1 32.61 3.6 16.57 3.6medium & low

Semi-arid temperate 12.5 1.5 24.0 20.8 11.8 16.2

Semi-Arid Tropic high 40.05 5.6 36.90 18.2 18.35 13.5Semi-Arid Tropic medium 35.15 17.4 30.76 10.2 17.65 10.9Semi-Arid Tropic low 42.38 26.3 35.54 12.1 18.13 13.6Semi-Arid Tropic 39.4 49.5 34.5 40.1 18.1 37.9

Arid high Neg Neg 5.77 0.6 0.95 0.1Arid low & medium 36.20 0.6 20.07 1.4 10.68 1.5Arid 33.4 0.6 19.0 3.3 9.9 2.2

All India 39.64 100.0 28.54 100.0 16.40 100.0(28.771) (40.323) (78.200)

Note: Figures in parenthesis indicate total population of the corresponding group in millions.Neg. = NegligibleSource: National Sample Survey 1999-2000

Appendixes

70

Overcoming Poverty in Rural India

Table 3.7. Percentage of poor among occupational groups.

Self employed in Non- Selfnon-agricultural Agricultural agricultural employed in Other

Agro-ecological zone activities labor labor agriculture occupations

Humid high 16.9 33.3 21.6 13.9 12.2Humid medium 17.0 31.9 31.6 15.9 8.3Humid low 18.0 41.4 25.2 20.3 10.6Humid 17.5 36.5 25.6 17.4 10.4

Semi-Arid Temperate high 14.4 28.0 16.0 8.2 11.5Semi-Arid Temperate medium & low 17.9 46.3 18.9 10.3 17.5Semi-Arid Temperate 14.9 31.1 16.5 8.6 12.2

Semi-Arid Tropic high 20.4 35.7 22.6 15.5 11.4Semi-Arid Tropic medium 17.7 36.5 19.7 15.1 12.4Semi-Arid Tropic low 17.2 38.1 27.6 18.3 7.4Semi-Arid Tropic 18.6 36.8 23.5 16.4 10.0

Arid high 1.2 6.1 5.4 1.0 0.7Arid low & medium 5.1 34.2 11.7 10.1 9.6Arid 3.5 17.4 9.5 6.5 5.6

All India 16.9 35.5 22.2 13.9 10.5

Source: National Sample Survey 1999–2000

Table 3.8. Incidence of poverty by farm size.

Agro-ecological zone Marginal Small Medium Large

Humid high 15.1 16.3 9.9 0.6Humid medium 14.9 18.3 17.7 7.4Humid low 16.8 23.6 23.3 24.0Humid 15.9 20.3 18.3 13.6

Semi-Arid Temperate high 11.4 7.8 3.7 1.4Semi-Arid Temperate medium & low 15.9 10.3 7.7 5.8Semi-Arid Temperate 12.1 8.2 4.3 2.1

Semi-Arid Tropic high 18.1 17.0 10.4 9.1Semi-Arid Tropic medium 20.2 13.5 16.8 10.4Semi-Arid Tropic low 22.8 19.2 20.4 12.4Semi-Arid Tropic 20.5 16.7 16.0 10.7

Arid high 0.0 0.8 0.1 2.0Arid low & medium 8.9 14.0 10.3 9.3Arid 8.2 13.0 9.5 8.7

All India 15.2 15.0 13.4 9.0

Source: ICRISAT Database 1998 and National Sample Survey 1999-2000

71

Table 3.9. Poverty among male and female headed households.

% of femaleMale Female All headed Sample

Agro-ecological zone headed headed households households size

Humid High 22.5 18.5 22.1 9.1 626Humid Medium 22.1 17.8 21.6 11.8 784Humid Low 26.3 20.1 25.6 11.3 1961Humid 24.3 19.2 23.7 10.9 3371

Semi-arid temperate High 13.0 21.0 13.7 8.2 875Semi-arid temperate medium & low 18.6 29.1 19.5 8.2 167Semi-arid temperate 13.9 22.2 14.6 8.2 1042

Semi-arid tropics High 23.5 28.6 24.0 10.3 788Semi-arid tropics Medium 22.9 26.0 23.2 8.7 578Semi-arid tropics Low 25.5 25.0 25.5 7.3 558Semi-arid tropics 24.1 26.8 24.3 8.8 1924

Arid High 2.6 5.6 2.8 5.1 63Arid medium & low 13.7 10.5 13.4 9.5 111Arid 12.9 9.1 12.6 7.3 174

India 21.2 21.8 21.3 10.3 6947

Source: National Sample Survey, 1999-2000

Table 3.10. Calories derived from various food items.

TotalPercentage of calories derived from

per capitaper day Meat Vege- Total

Agro- ecological zone (in Rs) Cereals Pulses Milk fish & egg tables Others non-cereals

Humid high 2145 71.2 4.4 4.8 1.2 6.2 12.2 28.8Humid medium 2148 74.2 3.8 3.5 1.1 6.7 10.7 25.8Humid low 2055 71.3 3.5 3.9 1.7 6.6 13 28.7Humid 2101.5 71.9 3.8 4.1 1.4 6.5 12.3 28.1

Semi-Arid Temperate high 2326 64.0 5.1 9.5 0.3 6.8 14.3 36.0Semi-Arid Temperate 2344 69.5 4.1 9.5 0.2 5.1 11.6 30.5medium and low

Semi-Arid Temperate 2328.9 64.9 4.9 9.5 0.3 6.5 13.9 35.1

Semi-Arid Tropic high 1958 69.4 4.6 5.9 0.7 5.5 13.9 30.6Semi-Arid Tropic medium 1985 67.5 4.7 7.1 0.5 4.7 15.5 32.5Semi-Arid Tropic low 2042 66.6 5.6 5.6 0.4 5.6 16.2 33.4Semi-Arid Tropic 1996.9 67.8 5.0 6.1 0.5 5.3 15.3 32.2

Arid high 2638 54.0 3.8 21.2 0.1 4.1 16.8 46.0Arid low & medium 2307 64.8 3.8 13.1 0.2 3.6 14.5 35.2Arid 2332.4 63.9 3.8 13.8 0.2 3.6 14.6 36.0

All India 2138 67.9 4.5 6.6 0.8 6.0 14.2 32.1

Source: National Sample Survey 1999-2000

Appendixes

72

Overcoming Poverty in Rural India

Table 3.11. Levels of expenditure and inequality.

Monthly per capita Expenditurefood expenditure Percentage of Price of cereals Inequality

Agro-ecological zone (Rs) MPCE* (Rs kg-1) (%)**

Humid high 284.9 61.7 8.1 30Humid medium 280.8 61.2 8.4 29Humid low 301.8 61.9 8.8 26Humid 292.3 61.7 8.5 28Semi-Arid Temperate high 291.1 56.8 7.1 27Semi-Arid Temperate medium & low 278.7 60.2 7.3 33Semi-Arid Temperate 289.1 57.4 7.1 28Semi-Arid Tropic high 285.3 59.4 7.7 28Semi-Arid Tropic medium 270.2 57.7 7.7 29Semi-Arid Tropic low 266.4 56.9 7.0 28Semi-Arid Tropic 273.8 58.0 7.4 28Arid high 410.4 55.5 6.5 28Arid low & medium 322.6 60.6 7.2 32Arid 329.3 60.1 7.1 32All India 287.5 59.2 7.9 29

* Share of food in total production** Ratio of bottom quintile to top quintileSource: National Sample Survey 1999-2000

Table 3.12. Wage rates (Rs day-1) in agriculture and non-agriculture.

Agricultural activities Non-agricultural activities

Agro-ecological zone Males Females Persons1 Males Females Persons

Humid high 39.5 30.4 36.5 69.8 38.1 65.6Humid medium 36.1 28.1 33.8 71.5 43.2 67.1Humid low 39.7 28.2 36.3 80.0 131.6 88.2Humid 38.8 28.8 35.8 75.2 85.2 77.0

Semi-Arid Temperate high 44.0 30.7 40.7 76.2 48.4 74.7Semi-Arid Temperate medium & low 35.1 28.9 33.2 63.7 52.6 62.2Semi-Arid Temperate 42.6 30.4 39.5 74.2 49.1 72.7

Semi-Arid Tropic high 42.9 30.2 37.4 74.1 110.3 81.3Semi-Arid Tropic medium 34.0 26.0 30.3 70.8 37.2 64.1Semi-Arid Tropic low 36.8 25.2 31.4 75.0 48.1 69.7Semi-Arid Tropic 37.9 27.1 33.0 73.4 65.2 71.8

Arid high 63.9 61.5 63.6 107.5 50.7 103.7Arid low & medium 33.8 24.0 28.4 60.8 41.3 56.5Arid 36.1 26.9 31.1 64.4 42.0 60.1

All India 39.2 28.0 35.0 75.5 74.1 75.3

1. Persons relate to both males and females

Source: National Sample Survey 1999-2000

73

Table 3.13. Children (6-14 years) participating in economic activities.

Agro-ecological zone Boys (%) Girls (%) Children (%)

Humid high 11.7 8.1 10.1Humid medium 7.6 5.2 6.4Humid low 7.7 7.1 7.4Humid 8.8 6.9 7.9

Semi-Arid Temperate high 6.6 5.0 5.8Semi-Arid Temperate medium & low 4.3 6.0 5.0Semi-Arid Temperate 6.2 5.2 5.7

Semi-Arid Tropic high 10.8 11.1 11.0Semi-Arid Tropic medium 13.8 15.9 14.8Semi-Arid Tropic low 13.6 12.5 13.1Semi-Arid Tropic 12.7 13.1 12.9

Arid high 10.3 2.5 6.7Arid low & medium 14.5 23.0 18.5Arid 14.2 21.4 17.6

All India 9.8 9.1 9.5

Source: National Sample Survey 1999-2000

Table 3.14. Work participation and unemployment rates.

Person day unemployment rate (%) Work participation rates (%)

Agro-ecological zone Males Females Persons1 Males Females Persons

Humid high 12.9 11.9 12.6 86.1 42.0 65.6Humid medium 12.0 11.5 11.9 83.1 36.8 61.2Humid low 11.8 11.1 11.6 83.9 44.2 65.1Humid 12.2 11.4 11.9 84.3 41.9 64.4

Semi-Arid Temperate high 5.2 2.4 4.5 83.6 37.4 61.8Semi-Arid Temperate medium & low 3.4 3.0 3.3 83.5 46.2 65.5Semi-Arid Temperate 4.9 2.5 4.3 83.6 38.8 62.4

Semi-Arid Tropic high 11.8 10.7 11.4 86.1 56.3 71.8Semi-Arid Tropic medium 5.9 5.4 5.7 87.1 64.3 76.0Semi-Arid Tropic low 7.2 6.9 7.1 86.2 62.8 74.9Semi-Arid Tropic 8.3 7.7 8.1 86.4 61.1 74.2

Arid high 5.4 1.0 4.3 83.5 47.8 66.2Arid low & medium 8.9 6.1 7.8 82.5 54.6 68.1Arid 8.6 5.7 7.5 82.6 54.1 68.0

All India 9.1 8.0 8.8 84.9 49.2 67.9

1.Persons relate to both males and females

Source: National Sample Survey 1999-2000

Appendixes

74

Overcoming Poverty in Rural India

Table 3.15. Percentage of children attending educational institutions.

Boys Girls Children Males Females Persons1

Agro-ecological zone (6-14 years) (6-14 years) (6-14 years) (15-19 years) (15-19 years) (15-19 years)

Humid high 70.4 62.9 66.8 44.8 30.6 38.3Humid medium 57.1 47.6 52.7 40.9 20.8 32.0Humid low 72.2 60.9 66.8 61.1 28.3 44.6Humid 68.3 58.4 63.6 51.9 27.2 40.0

Semi-Arid Temperate high 70.4 62.4 66.6 43.0 26.7 35.5Semi-Arid Temperate 67.6 57.6 62.9 52.1 27.8 41.3medium & low

Semi-Arid Temperate 69.9 61.6 66.0 44.5 26.9 36.4

Semi-Arid Tropic high 66.6 57.1 62.2 42.8 27.3 35.5Semi-Arid Tropic medium 66.0 53.5 60.4 42.5 17.4 31.3Semi-Arid Tropic low 64.9 49.4 57.6 49.1 34.3 41.5Semi-Arid Tropic 65.8 53.2 60.0 45.0 26.8 36.4

Arid high 61.7 49.1 55.8 39.6 20.8 31.2Arid low & medium 58.9 46.0 53.2 42.5 19.2 32.3Arid 59.1 46.2 53.4 42.3 19.3 32.2

All India 66.7 56.8 62.0 43.4 26.3 35.5

1. Persons relate to both males and femalesSource: National Sample Survey 1999-2000

Table 3.16. Primary and secondary education completion rates.

Primary in 12-14 years age group (%) Secondary in 15-19 age group (%)

Agro-ecological zone Males Females Persons1 Males Females Persons

Humid high 70.0 58.6 64.5 23.8 17.8 21.0Humid medium 54.3 41.4 48.8 13.2 10.1 11.9Humid low 58.7 47.3 52.6 19.5 16.4 18.0Humid 60.9 49.1 55.1 19.3 15.4 17.5

Semi-Arid Temperate high 65.9 58.0 62.0 22.4 18.1 20.4Semi-Arid Temperate medium & low 59.7 41.4 50.5 17.9 10.9 14.8Semi-Arid Temperate 64.9 55.3 60.1 21.7 16.9 19.5

Semi-Arid Tropic high 66.8 55.8 61.7 21.8 17.8 19.9Semi-Arid Tropic medium 54.9 44.7 50.2 16.4 9.1 13.1Semi-Arid Tropic low 51.7 43.7 48.1 20.4 13.0 16.6Semi-Arid Tropic 57.6 48.0 53.2 19.6 13.4 16.6

Arid high 57.0 43.9 51.1 19.9 14.3 17.4Arid low & medium 45.1 36.6 41.2 14.8 8.4 12.0Arid 46.0 37.2 42.0 15.2 8.9 12.4

All India 63.0 52.3 57.9 21.1 16.3 18.8

1. Persons relate to both males and femalesSource: National Sample Survey 1999–2000

75

Table 3.17. Number of dependents per worker.

Agro-ecological zone Young Old All

Humid high 0.9 0.1 1.6Humid medium 0.9 0.1 1.7Humid low 0.9 0.1 1.6Humid 0.9 0.1 1.6

Semi-Arid Temperate high 1.3 0.1 2.1Semi-Arid Temperate medium & low 1.0 0.1 1.6Semi-Arid Temperate 1.3 0.1 2.0

Semi-Arid Tropic high 0.7 0.1 1.1Semi-Arid Tropic medium 0.7 0.1 1.0Semi-Arid Tropic low 0.7 0.1 1.1Semi-Arid Tropic 0.7 0.1 1.1

Arid high 1.1 0.2 2.1Arid low & medium 0.8 0.1 1.3Arid 0.8 0.1 1.4

All India 0.9 0.1 1.5

Source: National Sample Survey 1999-2000

Table 3.18. Agricultural characteristics.

Fertilizer Use Crop Output Crop Output Livestock outputAgro-ecological zone (kg ha-1) per hectare (Rs) per capita (Rs) per capita (Rs)

Humid High 131 30505 3737 1107Humid Medium 61 22825 2816 906Humid Low 59 21070 2956 1349Humid 79 23995 3147 1173

Semi-Arid Temperate High 126 32926 4285 1333Semi-Arid Temperate Medium & Low 48 14435 3184 1195Semi-Arid Temperate 109 28284 4103 1310

Semi-Arid Tropic High 148 23314 3574 1290Semi-Arid Tropic Medium 80 15721 4118 1566Semi-Arid Tropic Low 63 13649 4637 1586Semi-Arid Tropic 85 16417 4137 1486

Arid High 117 34041 13273 3354Arid Low & Medium 37 9275 4894 1911Arid 70 17464 8251 2489

All India 86 20350 3946 1375

Source: ICRISAT Database 1998, National Sample Survey 1999-2000

Appendixes

76

Overcoming Poverty in Rural India

Table 3.19. Percentage share of public distribution system in total quantity purchased.

Agro-ecological zone Rice & wheat Sugar Kerosene

Humid high 13.8 48.1 71.4Humid medium 12.2 49.6 68.2Humid low 16.1 45.4 88.4Humid 14.6 47.1 79.0

Semi-Arid Temperate high 2.2 18.1 68.5Semi-Arid Temperate medium & low 1.7 27.8 75.1Semi-Arid Temperate 2.1 19.7 69.6

Semi-Arid Tropic high 13.3 51.1 70.6Semi-Arid Tropic medium 15.1 38.3 79.1Semi-Arid Tropic low 17.5 29.8 59.4Semi-Arid Tropic 15.4 39.4 69.1

Arid high 0.9 16.1 81.8Arid low & medium 17.0 29.9 77.3Arid 15.8 28.8 77.6

All India 12.4 34.8 76.5

Source: National Sample Survey 1999-2000

Table 3.20. Prices in the open market and public distribution system.

Percentage of PDS pricePDS price (Rs kg-1) to open market price

Rice& Rice &Agro-ecological zone wheat Sugar Kerosene wheat Sugar Kerosene

Humid high 4.1 10.4 4.0 42.7 65.0 41.7Humid medium 6.0 11.4 4.3 63.8 69.9 45.7Humid low 7.2 11.4 1.3 69.2 80.3 13.8Humid 6.1 11.1 2.7 60.9 73.2 29.0

Semi-Arid Temperate high 5.8 11.9 4.3 73.4 76.3 43.0Semi-Arid Temperate medium & low 4.0 12.2 4.2 50.0 75.3 47.2Semi-Arid Temperate 5.5 11.9 4.3 69.6 76.1 43.6

Semi-Arid Tropic high 4.0 11.9 3.8 71.4 75.8 39.6Semi-Arid Tropic medium 4.6 11.5 3.8 50.0 74.2 46.3Semi-Arid Tropic low 4.8 11.3 4.1 51.6 74.8 53.2Semi-Arid Tropic 4.5 11.6 3.9 55.6 75.0 46.1

Arid high 4.1 11.4 4.0 62.1 70.8 54.1Arid low & medium 5.1 11.8 3.9 66.2 71.1 58.2Arid 5.0 11.8 3.9 66.0 71.1 57.9

All India 5.3 11.5 2.9 62.4 74.7 31.9

Source: National Sample Survey 1999–2000

77

Table 3.21. Percentage of poor not purchasing and non-poor purchasing from public distribution system.

Percentage of poor not Percentage of non-poorpurchasing from PDS purchasing from PDS

Rice& Rice &Agro-ecological zone wheat Sugar Kerosene wheat Sugar Kerosene

Humid high 47.3 25.2 17.5 32.2 67.0 79.1Humid medium 49.9 21.3 17.6 32.2 65.3 81.1Humid low 49.4 21.4 18.8 37.8 71.8 81.5Humid 48.9 22.4 18.2 35.0 69.0 80.7

Semi-Arid Temperate high 63.6 42.8 16.3 4.5 41.4 72.3Semi-Arid Temperate medium & low 67.5 39.1 23.8 4.7 49.2 72.8Semi-Arid Temperate 64.2 42.2 17.5 4.5 42.7 72.4

Semi-Arid Tropic high 30.3 16.8 17.6 49.9 67.3 74.5Semi-Arid Tropic medium 42.1 19.4 15.0 38.8 70.7 79.0Semi-Arid Tropic low 40.4 28.1 33.3 41.4 61.7 63.0Semi-Arid Tropic 37.6 21.7 22.5 43.4 66.3 71.7

Arid high 80.3 56.2 15.3 2.9 65.3 72.9Arid low & medium 33.0 16.9 14.4 29.3 65.0 71.5Arid 36.6 19.9 14.5 27.3 65.0 71.6

All India 45.8 24.7 19.9 31.2 62.2 75.3

Source: National Sample Survey 1999-2000

Table 3.22. Coverage under anti-poverty programs during previous five years.

Agro-ecological zone IRDP (%) Employment (%)

Humid high 4.8 3.5Humid medium 5.2 3.1Humid low 6.4 3.8Humid 5.7 3.6

Semi-Arid Temperate high 5.7 2.4Semi-Arid Temperate medium & low 5.7 2.4Semi-Arid Temperate 5.7 2.4

Semi-Arid Tropic high 3.5 2.0Semi-Arid Tropic medium 5.5 2.8Semi-Arid Tropic low 5.7 2.5Semi-Arid Tropic 4.9 2.4

Arid high 6.9 1.3Arid low & medium 3.8 2.7Arid 4.0 2.6

All India 5.3 2.8

Source: National Sample Survey 1999–2000

Appendixes

78

Overcoming Poverty in Rural India

Table 3.23. Percentage of households using electricity for lighting and gas for cooking.

Agro-ecological zone Electricity For lighting Gas for cooking

Humid high 35.79 5.44Humid medium 24.89 3.70Humid low 39.39 5.49Humid 35.1 5.1

Semi-Arid Temperate high 36.48 5.25Semi-Arid Temperate medium & low 38.48 1.56Semi-Arid Temperate 36.8 4.7

Semi-Arid Tropic high 63.09 8.58Semi-Arid Tropic medium 66.96 5.10Semi-Arid Tropic low 69.27 5.91Semi-Arid Tropic 66.5 6.5

Arid high 81.12 10.74Arid low & medium 55.83 5.95Arid 57.8 6.3

All India 48.35 5.71

Source: National Sample Survey 1999–2000

Table 3.24. Computation of human development index for different zones.

Agro-ecological zone STDLIVING KNOWLEDGE HEALTH HDI

Humid High 0.253 0.466 0.416 0.378Semi-Arid Temperate High 0.287 0.453 0.360 0.367Semi-Arid Tropical High 0.249 0.434 0.454 0.379Arid High 0.394 0.371 0.499 0.421Humid Medium 0.240 0.352 0.375 0.322Semi-Arid Temperate Low and Medium 0.273 0.403 0.292 0.323Semi-Arid Tropical Medium 0.243 0.372 0.399 0.338Arid Low and Medium 0.306 0.318 0.398 0.341Humid Low 0.241 0.423 0.462 0.375Semi-Arid Tropical Low 0.249 0.389 0.472 0.370

Source: National Sample Survey 1999–2000

79

Tabl

e 3.

25. C

oeff

icie

nts

of c

orre

lati

on b

etw

een

pove

rty

and

soci

o ec

onom

ic v

aria

bles

.

H

umid

Hum

idH

umid

Sem

i-A

rid

Sem

i-A

rid

Tem

pSe

mi-

Ari

dIn

dia

Hig

hM

ediu

mL

owH

umid

Tem

p H

igh

Low

& M

ediu

m T

empe

rate

Cro

p ou

tput

per

cap

ita

–0.2

85**

*–0

.132

–0.0

430.

095

0.02

8–0

.548

***

–0.6

24**

*–0

.533

***

Cro

p ou

tput

per

hec

tare

–0.3

41**

*–0

.347

**–0

.433

***

–0.3

23**

*–0

.349

***

–0.4

40**

*0.

089

–0.3

92**

*Li

ve s

tock

out

put

–0.3

94**

*–0

.186

–0.1

60–0

.537

***

–0.3

70**

*–0

.476

***

–0.4

64**

*–0

.454

***

per

capi

taPe

rcen

tage

of

Fallo

w0.

204*

**0.

137

0.20

10.

130

0.15

6**

0.44

6***

0.42

00.

394*

**la

nds

Popu

lati

on d

ensi

ty–0

.025

–0.0

51–0

.231

–0.3

72**

*–0

.312

***

0.24

1***

0.38

20.

224*

*Fe

rtili

zer

per

hect

are

–0.2

89**

*–0

.303

0.08

9–0

.414

***

–0.3

28**

*–0

.446

***

0.23

1–0

.357

***

Prim

ary

educ

atio

n–0

.258

***

–0.0

19–0

.385

**–0

.721

**–0

.532

***

–0.2

79**

–0.3

76–0

.303

***

com

plet

ion

rate

Seco

ndar

y ed

ucat

ion

–0.2

45**

*–0

.192

–0.6

02**

*–0

.579

***

–0.5

17**

*–0

.114

–0.0

17–0

.107

com

plet

ion

rate

Lite

racy

rat

e–0

.321

***

–0.1

18–0

.568

***

–0.6

92**

*–0

.535

***

–0.4

19**

*–0

.094

–0.3

79**

*Pe

rcen

tage

of

SC–0

.003

0.22

70.

00–0

.086

–0.0

610.

338*

**0.

031

0.26

3*Po

pula

tion

Perc

enta

ge o

f ST

0.46

9***

–0.0

340.

581*

**0.

735*

**0.

630*

**–0

.157

–0.0

13–0

.040

Popu

lati

onPe

rcen

tage

of

Agr

i–0.

433*

**0.

446*

**0.

630*

**0.

635*

*0.

431*

**0.

628*

**0.

284

0.52

2***

cult

ural

Lab

orer

sPe

rcen

tage

of

Non

––0

.279

***

–0.3

65**

–0.4

21**

–0.4

69**

–0.3

92**

*–0

.407

***

0.00

3–0

.363

***

Agr

icul

tura

l Lab

orer

sW

ork

Part

icip

atio

n ra

tes

0.17

5***

0.15

50.

218

0.27

2***

0.25

9***

–0.0

01–0

.022

0.01

8Fe

mal

e w

ork

0.11

8**

0.00

40.

174

0.14

40.

150

0.03

30.

028

0.05

1Pa

rtic

ipat

ion

rate

Chi

ld w

ork

0.13

8***

–0.2

990.

168

0.52

3***

0.28

6***

0.07

8–0

.323

–0.0

61U

n–em

ploy

men

t ra

te0.

127*

*0.

228

–0.1

07–0

.052

–0.0

430.

170

0.08

10.

128

Mar

ket

dens

ity

–0.2

68**

*–0

.172

–0.2

10–0

.323

***

–0.2

55**

*–0

.349

***

–0.2

75–0

.337

***

Infr

astr

uctu

re I

ndex

–0.3

26**

*–0

.234

–0.1

89–0

.518

***

–0.3

74**

*–0

.343

***

–0.1

30–0

.305

***

Cer

eal P

rice

0.02

4–0

.053

–0.3

72**

–0.2

64**

–0.2

10**

*–0

.100

0.49

9**

–0.0

27A

gric

ultu

ral W

age

–0.4

64**

*–0

.486

***

–0.5

63**

*–0

.544

***

–0.5

08**

*–0

.468

***

–0.3

22–0

.453

***

Non

–Agr

icul

tura

l Wag

e–0

.266

***

–0.2

59–0

.288

–0.4

93**

*–0

.387

***

0.09

50.

075

0.09

6

Appendixes

80

Overcoming Poverty in Rural IndiaTa

ble

3.25

. Coe

ffic

ient

s of

cor

rela

tion

bet

wee

n po

vert

y an

d so

cioe

cono

mic

var

iabl

es (

cont

d...)

Sem

i-A

rid

Sem

i-A

rid

Sem

i-A

rid

Sem

i-A

rid

Ari

dA

rid

Trop

ic H

igh

Tro

pic

Med

ium

Trop

ic L

owTr

opic

Hig

h M

ediu

m &

Low

Ari

d

Cro

p ou

tput

per

cap

ita

–0.2

36–0

.336

**–0

.242

–0.2

49**

*–0

.006

0.01

5–0

.383

Cro

p ou

tput

per

hec

tare

–0.1

20–0

.315

**–0

.253

–0.1

96**

–0.2

77–0

.112

–0.4

38**

Live

sto

ck o

utpu

t pe

r ca

pita

–0.1

58–0

.353

**–0

.494

***

–0.3

27**

*–0

.081

–0.5

75–0

.464

**Pe

rcen

tage

of

Fallo

w la

nds

0.09

60.

156

0.17

30.

123

–0.0

450.

204

0.43

9**

Popu

lati

on d

ensi

ty0.

079

0.08

30.

137

0.06

9–0

.048

0.01

4–0

.017

Fert

ilize

r pe

r he

ctar

e–0

.032

0.01

5–0

.238

–0.8

20–0

.386

0.51

3–0

.201

Prim

ary

educ

atio

n–0

.045

–0.0

99–0

.129

–0.0

59–0

.355

0.46

90.

042

com

plet

ion

rate

Seco

ndar

y ed

ucat

ion

0.00

20.

217

–0.1

900.

010

–0.4

560.

138

–0.0

40co

mpl

etio

n ra

teLi

tera

cy r

ate

–0.2

88–0

.225

–0.1

39–0

.169

**–0

.622

–0.3

01–0

.452

**Pe

rcen

tage

of

SC P

opul

atio

n0.

370*

*0.

255

0.12

40.

231*

**0.

648*

*0.

205

–0.1

30Pe

rcen

tage

of

ST P

opul

atio

n–0

.076

0.12

60.

297*

*0.

142

–0.4

330.

282

0.34

0Pe

rcen

tage

of

Agr

icul

tura

l–0

.008

0.30

0**

0.34

6**

0.22

5***

0.14

50.

712*

**0.

445*

*L

abor

ers

Perc

enta

ge o

f N

on–A

gric

ultu

ral

–0.1

60–0

.397

***

–0.2

83**

–0.2

55**

*–0

.253

–0.2

91–0

.460

**L

abor

ers

Wor

k Pa

rtic

ipat

ion

rate

s–0

.125

0.07

7–0

.135

–0.0

860.

648*

*0.

334

0.28

7Fe

mal

e w

ork

Part

icip

atio

n ra

te–0

.134

0.07

4–0

.051

–0.0

630.

508

0.24

70.

264

Chi

ld w

ork

Part

icip

atio

n ra

te–0

.096

–0.1

730.

125

–0.0

520.

594

0.52

20.

618*

**U

n–em

ploy

men

t ra

te–0

.019

–0.0

040.

260

0.09

7–0

.120

0.60

4**

0.46

5**

Mar

ket

dens

ity

–0.2

83–0

.055

0.08

6–0

.096

0.16

80.

536

–0.2

35In

fras

truc

ture

Ind

ex–0

.095

–0.2

30–0

.195

–0.1

48–0

.650

**0.

029

–0.4

31**

Cer

eal P

rice

–0.2

190.

025

–0.1

39–0

.114

–0.7

54**

–0.1

19–0

.065

Agr

icul

tura

l Wag

e–0

.127

–0.3

57**

–0.4

92**

*–0

.332

***

0.10

0–0

.771

***

–0.6

31**

*N

on–A

gric

ultu

ral W

age

–0.1

32–0

.388

***

–0.1

75–0

.190

**–0

.279

–0.5

86**

–0.5

81**

*

**Si

gnif

ican

t at

5%

leve

l**

*Sig

nifi

cant

at

1% le

vel

Sour

ce: I

CR

ISA

T D

atab

ase.

199

8 an

d N

atio

nal S

ampl

e Su

rvey

199

9–20

00

81

Tabl

e 3.

26: R

esul

ts o

f re

gres

sion

equ

atio

ns –

Dep

ende

nt v

aria

ble:

inci

denc

e of

hea

d co

unt p

over

ty (

perc

enta

ge).

Hum

idSe

mi-

Ari

dSe

mi-

Ari

dSe

mi-

Ari

dSe

mi-

Ari

dSe

mi-

Ari

dV

aria

ble

Low

Tem

pera

te H

igh

Trop

ic M

ediu

m L

owH

umid

Tem

pera

te T

ropi

cA

rid

All

Indi

a

Con

stan

t16

8.1

94.3

124.

722

8.0

143.

997

.414

0.2

77.6

136.

6

Ln c

rop

–7.1

44–5

.975

**–5

4.57

**–8

.967

***

–8.2

95**

*–6

.624

***

Out

put

per

capi

ta(–

4.84

5)(–

2.06

9)(–

2.00

8)(–

3.98

2)(–

5.06

1)(–

3.76

9)

Ln c

rop

–3.1

99*

–18.

069*

**–3

.321

–5.2

43**

*–2

.915

***

Out

put

per

hect

are

(–1.

569)

(–2.

653)

(–1.

413)

(–3.

998)

(–2.

984)

Ln li

ve.

–19.

241*

**–6

.549

***

Out

putP

er c

apit

a(–

7.16

3)(–

4.69

9)Fa

llow

Lan

d (%

)Li

tera

cy–0

.220

***

–0.0

83–0

.064

Rat

e(–

2.65

9)(–

1.02

6)(–

1.42

1)ST

(%

)0.

420*

**0.

414*

**0.

305*

**(7

.314

)(7

.652

)(9

.119

)A

g. L

abor

(%

)0.

393*

1.80

2***

0.48

1*0.

348*

*0.

856*

**0.

382*

0.76

7***

0.32

4***

(1.7

37)

(5.6

45)

(1.7

27)

(1.9

75)

(2.7

38)

(1.8

69)

(4.0

32)

(2.9

54)

Non

-ag.

Lab

or (

%)

–0.1

61**

–0.4

13**

–0.2

63–0

.139

*–0

.176

*(–

1.90

1)(–

2.65

9)(–

1.27

2)(1

.524

)(–

1.55

3)W

ork

Part

icip

. Rat

e (%

)U

nem

ploy

. Rat

e (%

)0.

644*

*0.

359*

**(2

.633

)(3

.55)

Ln c

erea

l pri

ce–5

.161

Ln a

g. w

age

rate

–17.

644*

*–7

.739

**–5

.127

–11.

684*

*–1

1.87

2***

(–0.

832)

(–2.

463)

(–2.

010)

(–1.

440)

(–2.

031)

(–4.

851)

Ln n

on-a

g.–8

.964

***

–13.

142

–5.2

52–5

.884

*–7

.166

*–5

.873

***

Wag

e R

ate

(–2.

142)

(–1.

438)

(–1.

497)

(–1.

573)

(–1.

567)

(–2.

852)

R S

quar

e0.

809

0.65

70.

388

0.41

60.

617

0.53

90.

237

0.78

60.

526

F59

.973

***

31.0

89**

*6.

332*

**8.

559*

**35

.501

***

23.6

42**

*8.

312*

**15

.656

***

55.5

51**

*

Figu

res

in p

aren

thes

es a

re t

rat

ios

* Si

gnif

ican

t at

10%

leve

l**

Sig

nifi

cant

at

5% le

vel

***

Sign

ific

ant

at 1

% le

vel

Sour

ce: I

CR

ISA

T D

atab

ase,

199

8 an

d N

atio

nal S

ampl

e Su

rvey

199

9-20

00

Appendixes

82

Overcoming Poverty in Rural India

Table 3.27. Results of regression equations of crop productivity dependent variable: logarithm (cropoutput per hectare).

Semi-Arid Semi-AridVariable Humid Temperate Tropic Arid All India

Constant 8.900 8.281 7.969 7.400 8.495

Ln Fert. Per Hectare 0.078** 0.367*** 0.203*** 0.1390* 0.119***(2.202) (3.696) (4.287) (1.718) (4.556)

Irrigation Ratio 0.009*** 0.016*** 0.005*** 0.015*** 0.009***(4.825) (3.880) (3.900) (3.120) (9.930)

Literacy Rate 0.007*** 0.003 0.008*** 0.018*** 0.010***(4.002) (0.627) (3.811) (2.929) (7.496)

Ln. Market Density 0.040 –0.019*** 0.093** 0.075 0.018(1.143) (–2.921) (2.408) (0.745) (0.716)

R Square 0.312 0.577 0.511 0.909 0.455

F 17.6 27.2 34.8 40.0 83.8

Significance of F 0.0 0.0 0.0 0.0 0.0

Figures in parentheses are t ratios* Significant at 10% level

** Significant at 5% level*** Significant at 1% level

Source: ICRISAT Database 1998, and National Sample Survey 1999–2000

Appendix 4. Identifying Determinantsof Income Poverty – Framework

We try to elaborate how far the theory of productioncan help explain the existing level of income poverty,or alternatively, its opposite income growth andidentify its determinants. We know that profit or netincome from production of a commodity or provisionof a service (Y) can be expressed as follows:

Y = PQ – (R1X1+R2X2+ ----- +RnXn) ............. (4.1)

WhereY = Profit or net income from production

or provision of some service;P = Price of the commodity produced or

remuneration from the serviceprovided

Q = Quantity of produce or extent ofservice rendered;

R1, R2, --- Rn = Prices of inputs X1,X2, --- Xn; andX1, X2, --- Xn= Inputs used in the production of the

commodity or provision of theservice.

In words, this means that net income is the value ofthe commodity produced or the remuneration for theservice rendered less the cost of production of thecommodity or the provision of the service. We knowthat in perfectly competitive markets, prices of outputand inputs cannot be affected by the decisions of asingle producer/consumer. This means that a rationalproducer has to take the market prices of his produceand production inputs as given and then try tomaximize the production with a given bundle ofproduction inputs, or alternatively try to minimize thecost of producing a given level of output in order tomaximize his income. For this, the producer mustunderstand the nature and type of the underlyingproduction function and cost function. A typical farmproduction function could be expressed as follows:

Q = f (X1, X2, --- Xn) ...................................... (4.2)

Where Q is the quantity of farm output and X1, X2,---Xn are production inputs such as land, seeds,fertilizers, irrigation water, labor, mechanical andanimal power, machinery, management etc. Theseinputs together represent all the major categories offactors of production, ie, natural resources, humanresources, technology, capital, and management. Themodel could be used to estimate the nature and extent

of effect of various inputs on output of a given farmcommodity. When expressed in monetary terms, itcould also be used to estimate the effect of variousfactors on aggregate value of farm output. In that case,the dependent variable (Q) should be an index ofvalue of total farm output and the independent(explanatory) variables could also include measures ofinfrastructure such density of markets, orinfrastructure index. Shenngen et al. (2000) have usedthe Tornqvist-Theil index of agricultural output as thedependent variable in their model.

In the context of a rural household in SAT, we couldexpress them as a typical income generating functionas follows:

Y = Y fi + Ynfi ...................................................................................... (4.3)

WhereY = Household income in rupees per annum;Y fi =Income from farm enterprises; andYnfi = Income from non-farm sources.

In words, this means that the total householdincome is equal to the income from farm and non-farmenterprises or activities.

Given this formulation and the knowledge ofunderlying production functions, we could identifyvarious factors that affect household income. Thosefactors could be classified into the following categories1. Natural resources such as land, water (rainfall),

humidity, solar radiation, temperature and so on;2. Human resources including their knowledge, skills,

motivation, etc.;3. Technology comprising improved seeds, fertilizes,

pesticides, machinery, etc.4. Financial capital;5. Physical capital embodied in Infra-structure such as

roads, schools, hospitals, means of transport andcommunication, markets, etc.;

6. Organization and Management; and7. Public policies and programs.

Now, drawing upon the theory of production andbased on the knowledge of the underlying productionfunctions, we could formulate a set of hypothesesabout the possible relation between the level ofincome poverty or, alternatively, income growth and aparticular set of factors as specified above. Forexample, we could formulate a hypothesis that thelevel of farm income (poverty) in an area is affected,

83

84

Overcoming Poverty in Rural India

among other things, by the quantity and quality of landresources available – higher the per capita quantity andbetter the quality of land (in terms of productivity),ceteris paribus, higher the per hectare crop yield andhence higher the farm income, or lower the level offarm income poverty. This hypothesis could then betested empirically using appropriate statistical toolsand techniques.

We have used per capita household consumptionexpenditure as a proxy for per capita householdincome. This is because of the lack of availability ofestimates of household income for SAT and theavailability of household consumption estimates froma recent National Sample Survey (NSS) round. Theuse of household consumption expenditure as a proxyfor household income is universally recognized as asound practice as it is based on a relation betweenincome and consumption, We know that income isnormally used for consumption and saving, ie,

Y = C + S, or C = Y – S ............................... (4.4)

Where;Y = Household income;C = Household consumption expenditure; andS = Household savings.

So, from Equation 4.4, we know that, inequilibrium, household consumption is equal tohousehold income minus household savings. If S = 0,then, C = Y; if S > 0, then, C is less Y; and if S < 0,then, C is higher than Y. For a large sample from SAT,all these three conditions may hold and negativesavings may offset the positive savings, renderinghousehold consumption almost equal to household

income in the long run. From macro economic theoryof income determination, we know that

C = a + bY .................................................... (4.5)

In words, this means that consumption is a functionof some constant (threshold level of fixedexpenditure) and some fraction (b < 1) of income.This also establishes a definite relation betweenincome and consumption.

Based on the concepts, measures, and determinantsof income poverty presented in earlier paragraphs ofthis chapter, we have developed a conceptualframework connecting the context of poverty, causalfactors, interventions, and poverty reduction. It ispresented in Figure 2.1. This framework is intended toconnect logically the materials presented in subse-quent chapters of this report. An algebraic version ofthis framework could be expressed as follows:

P0i = f(FPi, Wi, LPi , IIi ,ALi ,STPI ....................... (4.6)

WhereP0i = The proportion of rural population that is poor

(Headcount index) for the ith district.FPi = Farm productivity index for the ith district.Wi = Wage rates for the ith district.LPi = Livestock Productivity Index for the ith

district.IIi = Infrastructure index for the ith district.ALi = Percentage of Agricultural Labor in the Rural

PopulationSTPI = Percentage of Scheduled Tribes in the Rural

Population

Appendix 5. Methodology for ComputingPoverty Indicators

The analysis is based on ungrouped household data ofthe 55th Round (1999-2000) of the National SampleSurvey (NSS) on Consumer Expenditure andEmployment and Unemployment. The sampleconsisted of 1,20,309 households (71,385 rural and48,924 urban) in the Consumer Expenditure Survey1,21,373 households (71,859 rural and 49,514 urban)in the Employment and Unemployment survey. Thehousehold data is aggregated for the ten agro-climaticzones using the household level multipliers developedby the National Sample Surveys Organization(NSSO). Initially, we started with 12 zones (fourbroad zones-Humid, Arid, Semi-Arid Temperate andSemi-Arid Tropical and each again divided into threeirrigation levels viz., less than 25% (low), 25 to 40%(medium) and above 40% (high). As our analysisfocuses on rural poverty, purely urban districts and afew districts for which data for classification were notavailable were not taken into account. Since thenumber of districts turned very small, low andmedium irrigation zones were clubbed together in caseof Arid and Semi-Arid Temperate zones, reducing thenumber of zones for analysis to ten. As validation forproper use of multipliers, the monthly per capitaexpenditures at the State level are first derived andcompared with the results published by the NSSO. Inaddition to these major sources, district-wise data onagricultural statistics of the Ministry of Agricultureand Cooperation and decennial population census datafrom Registrar General of Census operations are alsoused.

Poverty line is the most important parameterneeded in the estimation of poverty. The officialpoverty lines adopted by the Planning Commission for1999-2000 are available for each of the major states.Angus Deaton (2003) has also estimated poverty linesfor the same year on the basis of the price indicesderived from NSS consumption data. These ruralpoverty lines are lower than those adopted by thePlanning Commission in the official estimates for allthe states except Andhra Pradesh, Tamil Nadu,Karnataka and Gujarat (Appendix 3 Table 3.1) Deatonestimated poverty in India fusing these lines.However, he has adjusted the consumer expendituredata of the 55th Round for overstatement. As a resultof these two adjustments in opposite direction, hisestimate of rural poverty is very close to the official

estimate at 25.3%. We have adopted the poverty linesderived by Deaton, but no correction is made for theoverstatement of expenditure. As a result, the level ofrural poverty is found to be 21.3% in 1999-2000 ascompared to Deaton’s estimate of 25.3%. Furtherreduction in poverty is possible only throughinterventions in the high poverty areas andpredominantly poor socio-economic classes. Ruralpoverty is always related to the structure andperformance of agriculture. Time series analysisclearly shows positive association between povertydecline and agricultural growth. But the relationbetween rural poverty, agricultural development andagro-climatic conditions is not very clear.

The second step in the measurement of poverty isto choose the appropriate measure. Though HeadCount is the most popularly used measure, three othermeasures viz., Poverty Gap, Squared Poverty Gap andSen’s Index are also important for their properties.The first three measures belong to a class of additivemeasures. There are good surveys on themeasurement of poverty (Foster 1984; Atkinson1987). We briefly mention the main issues having abearing on policy analysis. Let y denote per capitaconsumer expenditure and z denote the poverty line.Let f(y) be the density function and F(y) be thecumulative distribution function (CDF). A functionf(y, z), non-increasing in y and non-decreasing in z is ameasure of poverty. A desirable property for thefunction is homogeneity. In other words, the measureis scale neutral. Various ways of aggregating the p(y,z)’s have been proposed in the literature. However,additive measures satisfy sub-group consistency,which means that when poverty increases in any sub-group of the population (say agricultural laborers)without a decrease elsewhere, the aggregate povertyshould also increase. A sub-group inconsistentmeasure may mislead policy analysis, as the measuremay not show decline in national poverty even when itdeclined in a particular area. The class of additivepoverty measures is given by

P (z) = ∫ p (y, z) f (y) dy .................................. (5.1)

The limits of integration are 0 and q. All the threemeasures of poverty viz., Head Count, Poverty Gapand Squared Poverty Gap are derived by taking(1-y/z)α for p (y, z) and giving 0,1 and 2 to α:

85

86

Overcoming Poverty in Rural India

P (z) = ∫ (1-y/z)α f (y) dy ................................ (5.2)

The limits of integration are 0 and q. The widely usedhead count index (H) is simply the proportion ofpopulation whose consumption (y) is less than thepoverty line (z). This is simply the value of P (z) whenα = 0 in equation (5.2). The measure is easy tounderstand and communicate, but it has two seriousdrawbacks which effect policy analysis. First, itviolates monotone axiom of welfare, which states thatan improvement in the income of some people, giventhe incomes of others, should reduce poverty. Headcount ratio is not sensitive to changes in income as longas these changes do not move a person from one sideof poverty line to the other. The measure also violatestransfer axiom of welfare – an idea first formulated byDalton (1920). The axiom states that transfers from aricher to a poorer person should reduce poverty. Thisviolation has a serious implication that a givenimprovement of incomes through policy interventionswill have high impact if those who are close to thepoverty line are selected.

The poverty gap index (PG) is obtained by settingα = 1 in equation (5.2). It measures the depth ofpoverty as it depends on the distances from thepoverty line as well as the number of poor. The widelyused income gap ratio is I = 1 – μP / z = PG / H,where, μP is the mean value of y for the poor. Itmeasures average proportionate shortfall below thepoverty line. This is a deceptive measure because if apoor person with a standard of living above μP escapespoverty, the income gap ratio will rise, though no one isworse off and one of the poor is, in fact, better off.Therefore, PG is a better measure than the incomegap ratio. While it satisfies the monotone axiom, it isinsensitive to transfers from a better off poor toanother poor person, as the gap remains the same aslong as both remain poor. While it gives depth ofpoverty, it does not indicate the severity of poverty, asit uses no weight for the gap from the poverty line. Thesquared poverty gap (SPG) index proposed by Foster-Greer-Thorbecke (1984) indicates severity of povertyand it is obtained by taking α = 2. This is a strictlyconvex function, a desirable property of a welfarefunction.

Sen (1976) proposed an index of poverty thatcombines the number of poor, the depth of poverty,and the distribution of the poor with in the group. Theformula is given by

Ps = 2/(q+1) n Σ (1–yi/z) (q+i+1) ................ (5.3)

Where q is the number of poor and q+i+1 is theweight accorded to the ith poor person from thepoverty line. The formula can be expressed in terms of

the average of the Head Count (P0) and Poverty Gap(P1) measures weighted by the Gini coefficient ofinequality among the poor (G p).

Ps = P0Gp + P1 (1 – G P) ................................ (5.4)

All the four measures of poverty are computed foreach zone and also for each district and State. Thedistrict level estimates of poverty are likely to sufferfrom high standard errors because of the small size ofthe sample at that level. Hence, they are used asobservations for the correlation and regression analysisto find the correlates and determinants of poverty.

The four measures of poverty viz., Head count,poverty gap, squared poverty gap and Sen’s Index arecomputed for each zone and also for each district andState. The district level estimates of poverty are likelyto suffer from high standard errors because of thesmall size of the sample at that level. They are,however, used as observations for the correlation andregression analysis to get an idea about the correlatesand determinants of poverty. Errors in dependentvariable will give unbiased estimates of theparameters, but the variances will be high. As a result,the regression coefficients are likely to becomeinsignificant. Errors in independent variables create amore serious problem as the parameters are unbiasedas well as inconsistent (Gujarati 1988). Errors invariables will not create a serious problem incorrelation analysis as both the variables underconsideration are assumed to be random.

Since poverty is a multi-dimensional concept, anattempt is also made to examine other aspects ofdeprivation such as education and health. Using theindicators of standard of living, knowledge and health,Human Development Index (HDI) is constructedadopting the methodology of United NationsDevelopment Programme (UNDP) (UNDP 2002).However, the choice of the variables is dictated by theavailability of data at the district level. The HDI iscomputed in two steps. In the first step, indices ofstandard of living, education and health areconstructed. The index of standard of living isconstructed using five variables viz., monthly percapita consumer expenditure, per capita calorieintake, inequality in consumer expenditure asmeasured by the ratio of per capita income of bottomquintile to top quintile of population, share of non-cereal food in total calorie intake and calorie intake ofthe bottom quintile of population. The index ofknowledge is measured by five variables viz., adultliteracy rate, participation of children in education,participation of the 15-19 age group persons ineducation, primary education completion rate among

87

children in the 12-14 age group and secondaryeducation completion rate among persons in the 15-19age group. For generating the index of health, childmortality and hospital and dispensary beds per lakhpopulation are adopted. After constructing the threebroad indices, they are combined into a single indexgiving equal weights. In this process, each healthindicator has got a higher weight than standard ofliving and knowledge indicators.

Before HDI is calculated, an index (dimensionindex) is created for each of the 12 variablesconsidered. To calculate these dimension indices,minimum and maximum values (goalposts) are chosenfor each underlying indicator from the district levelvalues.

actual value – minimum valueDimension index =

maximum value – minimum value

The standard of living index is calculated as thesimple average of the five dimension indices relating tothe standard of living. Similarly, the education index iscomputed by taking the simple average of the fivedimension indices relating to education. The health

index is the average of the two dimension indices ofhealth. The HDI index is the simple average of thestandard of living, education and health indices.

Further, the regression analysis for identifying thedeterminants of poverty is carried out only for 4 zonesfor want of sufficient degrees of freedom. In the caseof semi-Arid Temperate zone, separate regression wasworked out only for high-irrigated districts. Similarly, aseparate regression was estimated only for the lowirrigated districts of the Humid zone. In case of Semi-Arid Tropical zone, two separate regressions werefitted for low irrigated and medium irrigated sub-zones. In case of Arid zone, no separate regression wasworked out for any of the sub-zones. Regressionequations were estimated for the pooled data of thefour agro-climatic zones and for the country as awhole. Thus, nine regression equations wereestimated in all. In the regression equations thedependent variable was the Head Count ratio andsome of the independent variables were introduced inlogarithmic form. Regressions were also run for thefour zones and for the country to know about thedeterminants of crop productivity.

Appendixes

Appendix 6. A List of Selected Centrally-SponsoredRural Development Programs Currently

Underway in SAT, India

1. Natural resources development and managementprograms

Desert Development Programme (DDP);National Watershed Development Programmefor Dryland Areas (NWDPDA);National Wasteland Development Programme;Integrated Tribal Development Programme(ITDP); andMillion Wells Scheme;Afforestation, Social Forestry, and Joint ForestManagement.

2. Human resource developmentTraining of Youth for Self-Employment(TRYSEM);Development of Women and Children in RuralAreas (DWCRA);Integrated Child Development Programme(ICDP);Mid Day Meal Scheme;Scholarships for SC, ST and BC students;Reservation in government jobs for SC, ST andother BC persons; andAdult Literacy campaigns.

3. Programs for enhancing the productivity of cropsand livestock productivity

All India Coordinated Projects sponsored bythe Indian Council of Agricultural Research(ICAR);

Technology Missions – Oilseeds, Pulses, &Dairy Development;Operation Flood Programme

4. Programs for reducing risk and vulnerabilityCrop insurance;Livestock insurance;Drought relief works;Postponement of recovery of loans in times ofnatural calamities;National Old Age Pension SchemeNational Family Benefit Scheme

5. Poverty alleviation and employment generationSampoorna Grameen Rojgar Yojana (SGRY);Swarnajayanti Gram Swarozgar Yojana(SGSY)Employment Assurance Yojana (EAY)Jaiprakash Rozgar Guarantee Yojana (JPRGY)

6. Programs for strengthening basic infrastructurePradhan Mantri Gramodaya Yojana (PMGY)Central Rural Water and SanitationProgramme/Rajiv Gandhi Drinking WaterMission;Pradhan Mantri Gram Sadak Yojana;Rural Housing Schemes/Indira Awas Yojana(IAY);Rural Godowns Scheme; andRural Electrification Scheme.

88

Appendix 7. A Brief Critique of Some MajorPrograms and Lessons Learned

1. Natural resources developmentand management programsSoil conservation programs were the first in thiscategories to have been initiated in the fifties to reduceerosion and run off loses. The Drought Prone AreasProgramme (DPAP) was the first area developmentprogram launched by the Central Government in1973-74 to tackle the special problems engendered bydroughts. These areas are characterized by largehuman and cattle population, which continuouslyexert heavy pressure on the already fragile naturalresources base for food, fodder and fuel. The majorproblems are continuous depletion of vegetative cover,increase in soil erosion and fall in groundwater levelsbecause of continuous exploitation without any effortto recharge the underground aquifers.

The program aims at promoting the overalleconomic development and improving the socio-economic condition of the resource poor anddisadvantaged sections inhabiting the program areasthrough creation, widening and equitable distributionof resource base and increased employmentopportunities. The objectives of the program are beingaddressed in general by taking up development worksthrough watershed approach for land development,water resource development and afforestation/pasture development.

There are 183 districts spread over 16 states inIndia that are classified as drought prone districts.They are all covered under DPAP. All DPAP districtsare classified under four regions, namely, arid (withless than 375 mm annual rainfall); semi-arid (with 375to 750 mm rainfall); dry sub-humid (with 750 to 1125mm rainfall) and humid region (with rainfall of 1125mm and above). At present the DPAP is underimplementation in 971 blocks of 183 districts in 16States of India. The Desert DevelopmentProgrammes (DDPs) cover the arid areas of thecountry.

The DDP adopted the watershed approach in1987. The Integrated Wasteland DevelopmentProgramme launched by the National WastelandsDevelopment Board in 1989 also aimed at developingwastelands on watershed basis. The DesertDevelopment Programme focused on re-afforestationto arrest the growth of hot and cold deserts, whereas

DPAP concentrated on non-arable lands. Drainagelines for in-situ soil and moisture conservation, agro-forestry, pasture development, horticulture andalternate land use were its main components. TheIntegrated Wasteland Development Programme madesilvi-culture and soil and moisture conservation inlands under government or community or privatecontrol as its predominant activity.

In 1994, the Ministry of Rural Development,Government of India appointed a Technical Committeeon DPAP and DDP under the chairmanship of Dr.C.H. Hanumantha Rao to evaluate the implementa-tion and impact of these programs. The Committeerecommended that a common set of operationalguidelines, objectives, strategies and expenditurenorms for these three projects based on the watersheddevelopment approach should be evolved. The newguidelines, emphasizing community participation inthe program and capacity building, are now beingimplemented, as a follow up of the TechnicalCommittee’s recommendations.

An evaluation study of these programs has revealedthat with the sole exception of J&K, the revisedguidelines for implementing these programs are beingfollowed. The felt needs of the communities forphysical works are addressed. However, community’sactual involvement in planning and designing theseworks was generally weak. The condition of thephysical assets ranged from “poor” to “fair” wherethese were maintained by Watershed Associations /Committees or Panchayats except in Rajasthan andTamil Nadu, where the Community took ‘good’ careof these works. Government departments maintainedthese assets in Punjab and Manipur and their conditionwas good. However, the projects have succeeded inarousing community awareness of the need formanaging the land and water resources in an eco-friendly manner.

The impact of the program was perceptible both interms of increase in irrigation coverage, reduction insoil erosion and containing the process of soildegradation. Productivity increase in crops cultivatedwas also significant. In several States such asKarnataka, Uttar Pradesh, Kerala, Manipur, Meghalayathis impact was, however, not significant. There wasno perceptible shift in cropping patterns except inMaharashtra where horticulture crops were

89

90

Overcoming Poverty in Rural India

introduced. The impact of the program onencouragement of mixed farming or livestock relatedactivity was also not very significant. The impact onfarm incomes was positive but not significant except inStates such as Maharashtra, Punjab and Haryana.

One of the most striking features of the DPAP andother watershed-based schemes is the wide variationin quality of implementation (and of the assetscreated) which appears to correlate consistently withtype of implementing agency: government agenciesrarely have the time or skills to create (and help tomaintain) the degree of consensus which is necessaryfor strong local ownership of the resource. Non-Governmental Organizations (NGOs) generallyperform much better. The difficulty lies in the smallnumber of NGOs relative to the size of the task (andto the volume of disbursement of funds for watersheddevelopment) and the slow pace at which adequatelyskilled new ones can be created, though again someagencies have demonstrated foresight in training somemembers of watershed committees in rehabilitatedwatersheds to form NGOs which can act as ProjectImplementing Agencies (PIAs) in other nearbywatersheds (Farrington and Lobo 1997).

2. Human resourcesdevelopment programsSome important Human Resources DevelopmentProgrammes include Training of Youth for Self-Employment (TRYSEM), Development of Womenand Children in Rural Areas (DWACRA), NationalSocial Assistance Programme (NSAP), Mid-day MealScheme, scholarships for the Scheduled Castes (SC),the Scheduled Tribes (ST), and the Backward Castes(BC) students, Adult Literacy Campaigns, Jobreservations etc., However, these are not unique to therural people in SAT areas. They can avail theseschemes just as people from other regions can also do.

A TRYSEM evaluation study covering a sample of6686 beneficiaries arrived at e the followingconclusions: Of the total beneficiaries covered, around54% were women, which is more than the prescribednorm that 40% of the TRYSEM beneficiaries shouldbe women. Caste-wise distribution shows that 32% ofthe total beneficiaries belonged to SC and 21% to ST,which is as per the norms of the program (that 50%beneficiaries should be SC or ST). More than 85% ofthe sample beneficiaries were living in kaccha or semi-pucca houses. Around 70% of them were reported tobe from below poverty line (BPL) category. More thantwo-thirds of the beneficiaries trained were in the agegroup of 18 to 29 years.

Most of the states/UTs have not been able toachieve their physical targets in terms of the numberof youth to be trained. A look into the individualperformance of all the States/UTs reveals that in thereference period of three years, namely 1996 to 1999,Himachal Pradesh, Punjab, Gujarat and Haryana wereamong the top performers while Andaman & NicobarIslands, Maharashtra, Jammu & Kashmir and Biharperformed badly. While the reservation targets forwomen and SC/ST categories were achieved, thesame for handicapped (3%) has totally been neglected.There has been a 20% increase in the employmentlevels of the beneficiaries after training underTRYSEM. About 15% of them have started their ownemployment ventures. Only 26% of these receivedany subsidy/rebate/concession. More than 80% of theTRYSEM beneficiaries were satisfied with the qualityof training imparted and stipend offered to themduring training.

The DWCRA program was launched in 1982, aspart of the Integrated Rural Development Program(IRDP). Its aim was to empower rural women livingbelow the poverty line (BPL) by way of organizingthem to create sustainable income generating activitiesthrough self-employment. It was the first program ofits kind that specifically focused on improving thequality of life of rural women. A unique feature ofDWCRA, unlike other IRDP components, was thatalong with the improvement in income, it also focusedon access to health, education, safe drinking water,sanitation, nutrition etc., Thus it not only aimed ateconomic development, but also intended promotingsocial development. Another unique feature of theprogram was that it emphasized group activity. It wasthought that in the long run women’s empowermentdepends on creation of a movement that promotesawareness and self-reliance.

An evaluation study of DWCRA brought out thatDWCRA had a direct and significant impact onemployment and group activities. As high a proportionas 93% of the beneficiaries reported that DWCRA hadcreated a desire for self-employment. Whereas about89% of the beneficiaries felt that DWCRA had raisedtheir incomes. DWCRA had a visible impact onsavings, economic conditions and social prestige, but ithad less impact on health, sanitation, drinking waterand children’s education, which are more ofcommunity services. According to their overallperformance, the major states could be ranked (frombest to worst) as follows: Nagaland, ArunachalPradesh, Kerala, Andhra Pradesh, Assam, West Bengal,Mizoram, Tamil Nadu, Maharashtra, Orissa, Punjab,Karnataka, Madhya Pradesh, Tripura, Bihar, Jammu &Kashmir, Gujarat, Uttar Pradesh and Haryana.

91

The National Social Assistance Programme(NSAP), sponsored by the Ministry of RuralDevelopment (MoRD), Government of India, cameinto effect from August 15, 1995, with the purpose ofproviding social assistance to the rural poor in India. Itaims at improving the quality of life of the rural poorensuring equality and effective peoples’ participationin the process. The program extends 100% centralassistance to the States and Union Territories toprovide the benefits under it in accordance with thenorms, guidelines and conditions laid down by theCentral Government. The NSAP introduces anational policy for social security assistance to the poorfamilies in the case of old age, death of primarybreadwinner and maternity. It provides an opportunityfor linking the social assistance package to schemes forpoverty alleviation and provision of basic needs. Theprogram is being implemented through a synergisticpartnership with State Governments and under thedirect supervision of District Rural DevelopmentAgencies (DRDAs) in close collaboration with thevarious Panchayat Raj institutions. It has threecomponents, namely, National Old Age PensionScheme (NOAPS), National Family Benefit Scheme(NFBS) and National Maternity Benefit Scheme(NMBS).

Mid-day meal schemes are being implementedthrough out the country to improve enrollment ratiosand to supply nutritive food to the school children.Recently, Prime Minister has announced that thisscheme will be extended up to secondary school level.Scholarships and job reservations are beingimplemented to take care of the interests of sociallydisadvantaged groups. Adult Literacy campaigns arealso being implemented to improve the literacy levelsand participation of people in the developmentprograms.

3. Agricultural and livestockdevelopment programsA number of initiatives and programs were launchedby the government over the last five and half decadesbut they are not unique to SAT region. All IndiaCoordinated Research Projects were initiated on allmajor crop and live stock commodities to coordinatethe efforts of different research stations located allover the country and to evolve location specificproduction technologies for different regions.Technology missions were launched on pulses,oilseeds, maize and cotton, most of which are rainfedcrops. These missions strived to further intensifyresearch and gave fillip to technology adoption by

removing some bottlenecks and constraints. The AllIndia Coordinated Research Project for DrylandAgriculture tried to identify efficient crops andvarieties suitable to the growing seasons of differentdryland zones in the country. While all these programscontributed to the agricultural and livestockdevelopment of the country in general and drylandregions in particular, the technological and productionconstraints are still constraining the growth ofagricultural and livestock development in the SATregions.

Operation Flood is the world’s biggest dairydevelopment program. It was designed and launchedin 1970 by the National Dairy Development Board,then a non-governmental organization but now astatutory corporate body. The program sought toreplicate the Anand Pattern dairy co-operativestructure in selected milk sheds in India. The AnandPattern dairy co-operatives formulate and implementtheir own policies and programs for dairydevelopment in their area and hire professionalmanagers and technicians for their implementation.The role of the government is limited to assisting theco-operatives financially in implementing their ownprograms. The government funds for dairydevelopment are placed at the disposal of the co-operatives.

Of is a living example of a non-governmentalsystem that has been successful in promoting people-centered development more efficiently and effectivelythan the governmental system. OF has enabled Indiato emerge as the world’s highest milk producingcountry. Evaluation studies of OF show that theprogram has significantly improved the income of ruralmilk producers, especially marginal and small farmersand landless laborers (Singh 1999). OF has greatpotential for replication in many SAT areas, wherelivestock production is catching on.

4. Wage paid employmentgenerating schemesBoth the Swarnajayanti Gram Swarozgar Yojana(SGRY, formerly Food for Work) and EmploymentAssurance Scheme (EAS) aim to provide additionalwage employment in rural areas, particularly during‘lean’ seasons. Among the differences between themare the stronger emphasis on paying part of the dailywage in the form of food grains under the SGRY, and astronger provision for materials to allow the creationof durable assets under the EAS (Nayak et al. 2000).Both schemes are to a large degree self-targeting, sinceonly the poorest will work for the basic minimum

Appendixes

92

Overcoming Poverty in Rural India

wage. The strong role played by ‘payments in kind’ –via food grains – in the case of SGRY, and the cost anddifficulty of storing and transporting grains, meansthat there is little scope for local-level misappro-priation of resources by officials. However, within-district responsibilities for storage, transport anddistribution of food grains are contracted out to fairprice shops and private contractors, and there isconsiderable scope here for fraudulent practices when(as until recently has been the case) the publicdistribution price is lower than the market price. Theuse of food grains for payment in kind is minimal in theEAS, and EAS regulations prohibit the engagement ofcontractors or middlemen, but this provision is widelyflouted, and has been accompanied by falsification ofattendance lists and other irregularities. Someestimates suggest that, given the combination ofmalpractice among administrators and contractors,only 25% of the wage funds to which beneficiaries areentitled actually reaches them, the remaining 75%disappearing through leakages of various kinds.

5. Self-employment generatingschemesThe Integrated Rural Development Programme(IRDP) is one of the largest and longest-standingefforts towards self-employment generation in Indiaand is implemented by government agencies. Bycontrast, the National Credit Fund for Women,popularly known as Rashtriya Mahila Kosh (RMK) isrelatively new, still small, and implemented throughNGOs. The IRDP operates through a mixture ofsubsidy and bank loan. The subsidy element has beensubstantial, in the range of Rs 4000 to Rs 6000depending on beneficiary characteristics. As with theNational Housing Scheme, subsidies of this magnitudehave attracted the interest of politicians wishing todivert the subsidies to their current or potentialsupporters. This has contributed to low repaymentrates on the loan component, since defaulters (andbank staff) are aware of the political support enjoyedby this category of beneficiaries. It has also meant thata high proportion of beneficiaries are not below thepoverty line.

Malpractice by lower-level officials has beenpervasive. Surveys in some areas indicate that a 10%

deduction was made by bank officials as informal‘charges’. In other localities, over 20% of the subsidycomponent was charged in various ways as ‘speedmoney’. Another common form of corruption in someareas was for officials in collusion with local middle-men to provide the asset specified by beneficiaries,contrary to the regulations which require these to beprovided by approved suppliers in exchange for cashpayments by the beneficiaries. Working in collusionwith administrators, the banks have also made illicit‘charges’ on beneficiaries.

There is a large literature on the limitations of theIRDP in helping to create a sustainable productiveasset base for the low-income self-employed. Difficul-ties include the fact that the production systems forassets specified for investment under the IRDP are notsupported by services available through other relevantgovernment departments. In addition, in many casesthere is no insurance provision for assets such aslivestock, so that their death will make it impossiblefor beneficiaries to repay loans. Other problemsinclude the limited capacity of government to identifyinvestment opportunities (resulting in excessiveinvestment in one particular kind of asset – such asdairy cows – within limited areas, so that marketsquickly become flooded); Frequent loan moratoriainstigated by politicians, so that few beneficiaries takeseriously the requirement to repay; Inadequatemonitoring at activity or output levels so thatproblems of inappropriate assets, high risk levels andlow returns are rarely detected; and Complexapplication procedures, which are near impossible forlow-income people to understand.

By contrast, the RMK focuses entirely on womenbelow the poverty line and uses minimal procedures todisburse loans through NGOs. These then arerequired to identify women’s groups eligible for loansand to contribute a 10% margin on the loan. There isno subsidy component. Funds disbursed through theRMK attract very little political attention, largelybecause of the absence of any subsidy component.Government officials have very little to do withdisbursement procedures, and little corruption occurs.Evaluations indicate the substantial contribution madeto women’s livelihoods by the assets purchased.However, the limited number of NGOs suitable forimplementing this scheme is likely to constrain itsexpansion severely.

Citation: Rao KPC, Bantilan MCS, Singh K, Subrahmanyam S, Deshingkar P, Rao P Parthasarathy and Shiferaw B.2005. Overcoming Poverty in Rural India: Focus on Rainfed Semi-Arid Tropics. Patancheru 502 324 AndhraPradesh, India: International Crops Research Institute for the Semi-Arid Tropics. 96 pp. ISBN 92-9066-474-6.Order code BOE 035.

Abstract

This study is a comprehensive analysis of the nature, the dynamics and the determinants of poverty in thesemi-arid tropics (SAT). It uses the sustainable livelihood framework to identify the underlying determinantsof poverty and the pathways in moving out of poverty. Quantitative analysis based on the most recent NationalSample Surveys (NSS) of India (1999–2000 round) generated classifications of relevant typologies includingagroecological zones, occupational groups, irrigation access and social groups. It shows that poverty in India isconcentrated most in the SAT and humid areas. It identifies caste and occupational groups facing highestincidence of poverty. The typologies systematized the establishment of linkages between rural poverty and thesemi-arid tropics.

The study presents a synthesis of evidence and lessons learnt from the village level studies (VLS) by theInternational Crops Research Institute for the Semi-Arid Tropics (ICRISAT) and other household level casestudies – income and expenditure levels, incidence of poverty and under-nutrition and coping strategies whenfaced with droughts and other income shocks. The empirical analysis uses both the quantitative and thequalitative data to document the existing and emerging evidences on the spatial and the social dimensions ofpoverty. The poverty diagnostics and strategic assessment of the issues confronting SAT generated acomprehensive perspective from both the micro and the macro levels, thus deepening our understanding ofthe multi-dimensional nature of SAT poverty.

The study reviews the existing poverty alleviation programs (their merits and demerits and successes andfailures) and concludes with a set of recommendations on strategies to increase the contribution of SATagriculture to poverty reduction.

Résumé

La présente étude fait une analyse détaillée de la nature, de la dynamique et des déterminants de la pauvretédans les zones tropicales semi-arides (SAT). Elle se s’appuie sur le cadre de moyens de subsistance durablespour identifier les facteurs qui déterminent la pauvreté et explorer les pistes qui permettent d’en sortir. Uneanalyse quantitative, basée sur les enquêtes les plus récentes faites en Inde (1999-2000), à partir d’échantillonsnationaux (National Sample Surveys – NSS), s’est traduite par des classifications pertinentes - zones agro-écologiques, groupes professionnels, accès à l’irrigation et groupes sociaux. Cette analyse montre, qu’en Inde,la pauvreté se concentre surtout dans les SAT et les zones humides. Elle permet d’identifier les castes et lesgroupes professionnels qui subissent le plus les conséquences de la pauvreté. Les typologies ont systématisél’établissement de liens entre la pauvreté en milieu rural et les zones tropicales semi-arides.

L’article résume les preuves et les enseignements issus des études menées au niveau villageois (VLS) parl’Institut international de recherche sur les cultures des zones tropicales semi-arides (ICRISAT) et d’autresétudes de cas réalisées au niveau des ménages – niveaux des revenus et des dépenses; incidence de la pauvreté,et sous-alimentation et stratégies d’adaptation en cas de sécheresse et autres contrecoups liés au revenu.L’analyse empirique s’appuie à la fois sur les données quantitatives et qualitatives en vue d’étayer les preuvesexistantes et émergentes, relatives aux dimensions spatiales et sociales de la pauvreté. Le diagnostic de lapauvreté et l’évaluation stratégique des problèmes que connaissent les zones tropicales semi-arides, ont permisd’avoir une perspective générale, d’un point de vue à la fois micro et macro, renforçant ainsi notre compréhensionde la nature multidimensionnelle de la pauvreté dans les SAT.

L’article examine les programmes qui existent en matière d’atténuation de la pauvreté (avantages/inconvénients;succès et échecs) et se termine par une série de recommandations sur les stratégies qui permettent d’améliorerla contribution de l’agriculture dans les SAT à la réduction de la pauvreté

ISBN 92-9066-474-6 Order code BOE 035 614–2004

ICRISAT-Patancheru(Headquarters)Patancheru 502 324Andhra Pradesh, IndiaTel +91 40 23296161Fax +91 40 23241239icrisat@cgiar.org

ICRISAT-Niamey(Regional hub WCA)BP 12404Niamey, Niger (Via Paris)Tel +227 722529, 722725Fax +227 734329icrisatsc@cgiar.org

ICRISAT-Nairobi(Regional hub ESA)PO Box 39063, Nairobi, KenyaTel +254 20 524555Fax +254 20 524001icrisat-nairobi@cgiar.org

ICRISAT-BulawayoMatopos Research StationPO Box 776,Bulawayo, ZimbabweTel +263 83 8311-15Fax +263 83 8253/8307icrisatzw@cgiar.org

ICRISAT-BamakoBP 320Bamako, MaliTel +223 2223375Fax +223 2228683icrisat-w-mali@cgiar.org

ICRISAT-Maputoc/o INIA, Av. das FPLM No 2698Caixa Postal 1906Maputo, MozambiqueTel +258-1-461657Fax +258-1-461581icrisatmoz@panintra.com

ICRISAT-LilongweChitedze Agricultural Research StationPO Box 1096Lilongwe, MalawiTel +265-1-707297/071/067/057Fax +265-1-707298icrisat-malawi@cgiar.org

The International Crops Research Institute for the Semi-Arid Tropics (ICRISAT) is a non-profit, non-political, international organization for science-based agricultural development. ICRISAT conductsresearch on sorghum, pearl millet, chickpea, pigeonpea and groundnut – crops that support the livelihoodsof the poorest of the poor in the semi-arid tropics encompassing 48 countries. ICRISAT also sharesinformation and knowledge through capacity building, publications, and information and communicationtechnologies (ICTs). Established in 1972, it is one of 15 Centers supported by the Consultative Group onInternational Agricultural Research (CGIAR).

Contact information

About ICRISAT

Visit us at www.icrisat.org

Overco

min

g P

overty in

Ru

ral Ind

ia