the rural non-farm economy: prospects for growth and poverty reduction

World Development Vol. 38, No. 10, pp. 1429–1441, 2010� 2010 Elsevier Ltd. All rights reserved

0305-750X/$ - see front matter

www.elsevier.com/locate/worlddevdoi:10.1016/j.worlddev.2009.06.008

The Rural Non-farm Economy: Prospects for Growth

and Poverty Reduction

STEVEN HAGGBLADEMichigan State University, USA

PETER HAZELLUniversity of London, UK

and

THOMAS REARDON *

Michigan State University, USA

Summary. — Policy makers have high expectations for the rural non-farm economy (RNFE). Given high income shares, growingemployment, and frequently low capital requirements, they see the RNFE as a potential pathway out of poverty for their rural poor.Yet available evidence suggests that pro-poor rural non-farm growth does not occur automatically. For the poor to benefit from ruralnon-farm growth, policy makers must stimulate buoyant rural economies, with robust non-farm income growth, not simply low-produc-tivity employment. Moreover, the poor must gain access to growing market niches. Fluid labor markets provide one important bridgelinking the rural poor to growing non-farm opportunities.� 2010 Elsevier Ltd. All rights reserved.

Key words — non-farm, rural, poverty, inequality

�Final revision accepted: June 23, 2009.

1. INTRODUCTION

Non-farm earnings account for 35–50% of rural householdincome across the developing world (Table 1). Landless andnear-landless households everywhere depend heavily on non-farm income for their survival, while agricultural householdscount on non-farm earnings to diversify risk, moderate sea-sonal income swings, and finance agricultural input purchases.Over time, the rural non-farm economy has grown rapidly,contributing significantly to both employment and rural in-come growth.

Long neglected by policy makers, the rural non-farm econ-omy (RNFE) has attracted considerable attention in recentyears. In poor agrarian countries, struggling with growingnumbers of marginal farmers and lackluster agricultural per-formance, such as much of Africa, policy makers view theRNFE as a potential alternative to agriculture for stimulatingrural income growth. In successfully transforming economies,they see the RNFE as a sector that can productively absorbthe many agricultural workers and small farmers beingsqueezed out of agriculture by increasingly commercializedand capital intensive modes of farming. Given frequentlylow capital requirements in the non-farm economy, policymakers in both settings view the RNFE as offering a potentialpathway out of poverty for many of their rural poor. Expecta-tions everywhere are high.

How realistic are these expectations? Can the RNFE indeedgrow rapidly enough to productively absorb a growing rurallabor force? And in doing so, can it, in fact, provide a pathwayout of poverty for the rural poor? Affirmative answers to thesequestions require two preconditions: that suitable engines ofgrowth are available to drive a buoyant RNFE and that thepoor control sufficient assets of the right kind to enable them

1429

to access the new non-farm income earning opportunities thatarise. Satisfying these conditions, in turn, depends crucially onthe specific economic and social context in which the RNFEoperate, for across the developing world, diversity abounds.Therefore, opportunities and appropriate policy stances differsignificantly across settings. This paper examines key factorsaffecting growth and equity in the rural non-farm economyin order to identify settings and policies favoring pro-poor rur-al non-farm growth.

The paper begins with some key definitions followed by adescriptive profile of the rural non-farm economy. Discussionthen moves into rural dynamics, exploring internal processesof rural growth and change as well as the impact of growingglobalization and urbanization on rural non-farm activity.The paper then examines the conditions under which thetwo preconditions for pro-poor growth of the RNFE can bemet.

2. DEFINITIONS

Following standard convention, this paper considers thenon-farm economy to include all economic activities other thanthe production of primary agricultural commodities. Non-farm, thus, includes mining, manufacturing, utilities, construc-tion, commerce, transport, and a full gamut of financial,personal, and government services. Agroprocessing—the trans-formation of raw agricultural products by milling, packaging,bulking, or transporting—forms a key component of the ruralnon-farm economy. Likewise, following a long tradition, 1 this

http://dx.doi.org/10.1016/j.worlddev.2009.06.008

Table 1. Non-farm share of rural income. Source: Reardon et al. (2007),Table 6.1, summarizing 54 rural income surveys from the 1990s and 2000s

Region Non-farm share of rural income

Total non-farmearnings (%)

Local non-farmbusiness and

employment (%)

Transfersand remittances

(%)

AfricaExcludingNamibia

34 28 6

Namibia 75 26 48

Asia 51 40 11Latin America 47 41 6

Total citations include 23 for Africa, 2 for Namibia, 14 in Asia, and 17 forLatin America.

1430 WORLD DEVELOPMENT

paper adopts a broad definition of rural regions as encompass-ing both dispersed rural settlements as well as the functionallylinked rural towns where many agroprocessing and ancillarynon-farm service and commercial activities congregate to ser-vice surrounding agricultural settlements.

In measuring income, the classification of migrant remit-tances and income transfers is less standardized. Generally,earnings from commuting or temporary and seasonal migra-tion by family members who remain part of a rural householdare included as rural non-farm income. However, transfers bygovernment, former household members, or relatives are oftenclassified separately as transfer income. Many studies, how-ever, fail to distinguish clearly among these various sourcesof non-farm income. To help establish rough orders of magni-tude, Table 1 breaks out locally earned non-farm business andwage income from remittances and transfers.

3. CHARACTERISTICS OF THE RURAL NON-FARMECONOMY

(a) Size

Policy interest in the RNFE arises in large part because ofits increasing importance as a source of income and employ-ment across the developing world. Primary employment data,which offer the most widely available indicator of the scale ofrural non-farm activity, suggest that the RNFE accounts forabout 30% of full-time rural employment in Asia and LatinAmerica, 20% in West Asia and North Africa (WANA), and10% in Africa (Table 2). Inclusion of rural towns—which

Table 2. Composition of rural non-farm employment, by region (%). Source: 31

Non-farm shareof rural workforce

Women’s shareof rural non-farm

employmentManufacturing

Africa 9 39 19Asia 24 24 27Latin America 31 36 22West Asia andNorth Africa

21 11 23

Regional aggregates weighted by size of total primary workforce. Country daa Trade and transport includes wholesale and retail trade, transport, and storab Other services includes finance, insurance, and community and social servicec Other includes mining and quarrying, utilities, construction, and other non-c

frequently depend on the rural hinterlands for both inputsand markets—raises non-farm employment shares by an addi-tional 10–15% (Hazell & Haggblade, 1993). Because theseemployment data typically measure only primary occupations,to the exclusion of secondary and seasonal pursuits, they oftenunderstate the importance of rural non-farm activity.

Income data, which include earnings from seasonal andpart-time activity, offer a more complete picture of the scaleof the RNFE. Evidence from a wide array of rural householdsurveys suggests that non-farm income accounts for about35% of rural income in Africa and roughly 50% in Asia andLatin America (Table 1). Standing roughly 20% higher thanthe comparable employment data, these income shares con-firm the economic importance of part-time and seasonalnon-farm activities.

Rural non-farm employment holds special importance forwomen. Women account for about one-quarter of the totalfull-time RNFE workforce in most parts of the developingworld (Table 2). Given their frequently heavy household obli-gations and more limited mobility, women also participate inpart-time RNFE activity, particularly in household-basedmanufacturing and service activities.

Likewise, rural landless and near-landless households de-pend heavily on non-farm income sources. Those with lessthan 0.5 ha earn between 30% and 90% of their income fromnon-farm activities (Hazell & Haggblade, 1993).

(b) Composition

The rural non-farm economy houses a highly heterogeneouscollection of trading, agroprocessing, manufacturing, com-mercial, and service activities. Even within the same country,strong differences emerge regionally, as a result of differingnatural resource endowments, labor supply, location, infra-structural investments, and culture (Table 3).

The scale of individual rural non-farm businesses variesenormously, from part-time self-employment in household-based cottage industries to large-scale agroprocessing andwarehousing facilities operated by large multinational firms.Often highly seasonal, rural non-farm activity fluctuates withthe availability of agricultural raw materials and in rhythmwith household labor and financial flows between farm andnon-farm activities (Figure 1).

Sectorally, despite a common policy emphasis on ruralindustries, manufacturing typically accounts for only 20–25% of rural non-farm employment, while trade, transport,construction, and other services account for 75–80% (Table2). Spatially, the composition of non-farm activity varies as

population censuses as summarized by Haggblade et al. (2007), Table 1.2

Rural non-farm employment shares

Trade andtransporta

Financial andpersonal servicesb

Construction, utilities,mining, and otherc

Total ruralnon-farm

31 35 15 10029 31 14 10023 34 21 10022 36 20 100

ta weighted by size of total primary work force.ge.slassified activity.

Table 3. Heterogeneity of rural non-farm activity (percent of total rural employment). Source: Ferreira and Lanjouw (2001), Lanjouw (2001), andLivingstone (1997, chap. 12)

Primary employmenta Rural Brazil, 1996 Rural El Salvador, 1994 Rural Uganda, 1992

Northeast Southeast West Central East Hoima Mukono

Mining and natl resources 5.5 1.1 0 0.9 0.3Manufacturing 3.5 5.6 10.6 13.4 5.6 15.9 13.6

Beverages 0.1 0.2 11.2 3.3Food processing 1.1 1.3Textiles 0.6 0.5 2 3.8 1.5 1.2 0Straw 0.6 0.7 2.3 3.1 0.6 2.3 8.6Wood 0.6 0.5 1.2 0.7Shoes and leatherware 0.1 0.1 0 1Other manufacturing 0.4 2.3 6.3 6.5 3.5

Construction 2.6 4 4.1 6.5 2.8 5.1 5.3Utilities 0.2 0.5 0.4 0.4 0Commerce 4.4 4.6 9.3 12.7 9 32.1 25.3

Retailing and wholesaling 3.7 3.2 7.1 9.9 7.1 32.1 25.3Transport 0.7 1.4 2.2 2.8 1.9

Services 10.8 16.5 9.2 14.6 5.5 7.1 16.8Restaurant/hotel 0.8 1.3 0.1 1.3 0 2 8.8Finance 0.1 0.2 0.2 0.2 0.2Teaching 2.9 2.2 0.8 0.4 0.9Health 0.5 0.5 0.2 2.3 1Domestic service 3 5.8 2Personal services 1 0.9Government 1.4 1.3 0.5 0.1 0.1Other services 4.6 10.6 4.1 6.3 2.1 2.8 7

Total non-farm 27 32.3 33.6 48.5 23.2 60.2 61Total agriculture 73 67.7 66.4 51.5 76.8 – –Total employmenta 100 100 100 100 100 – –

a Uganda data list percent of households involved. All others are percent of primary employment.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Jan

Feb

Marc

hApril

May

June Ju

lyAug

Sept

OctNov

Dec

perc

ent o

f la

bor

days

by

activ

ity

agriculture nonfarm

Figure 1. Seasonality of rural non-farm employment, Ethiopia 1993.

Distribution of employment among landless laborers. Source: Habtu

(1997, chap. 3).

THE RURAL NON-FARM ECONOMY: PROSPECTS FOR GROWTH AND POVERTY REDUCTION 1431

well. While home-based cottage industries predominate in rur-al areas, towns and urban centers support an increasing con-centration of factory manufacturing, services, and trade(Table 4).

The composition of services differs as well. While rural areashouse small retailers, basic farm equipment repair services,and input supply firms, other services such as primary school-ing, health clinics, barber shops, milling, and transport facili-ties tend to locate in small towns. Larger settlements attractcinemas, restaurants, wholesale distributors, and higher-levelschool, health, and communication facilities (Wanmali, 1983).

Government services, likewise, provide significant ruralemployment opportunities in some instances. In rural Egypt,government employment generates over 45% of rural non-farm incomes, while in rural Pakistan, government jobs pro-vide about 25% of rural non-farm earnings (Adams, 2002;Adams & He, 1995). In rural India, government jobs accountfor nearly 20% of rural non-farm employment (Fisher, Maha-jan, & Singha, 1997).

Remittances account for a large share of rural income insome locations. In the mining economies of Southern Africa,remittances may account for as much as half of all ruralhousehold income. They likewise form an important part ofhousehold income diversification and risk reduction strategies(Barrett, Reardon, & Webb, 2001; Ellis, 2000; Ellis & Free-man, 2004). In most rural settings, however, local businessand wage income account for a majority of non-farm earnings,while remittances and transfers typically account for 15–20%of non-agricultural rural income and 5–10% of total rural in-come (Table 1).

(c) Equity implications

The extreme heterogeneity of rural non-farm activity resultsin widely varying productivity and profitability (Table 5). Re-turns vary substantially, normally as a function of differingphysical and human capital requirements. Women dominatemany of the low-return cottage industries, while the poordominate other low-return activities, such as small-scale trad-ing and unskilled wage labor used in construction, portering,and many personal services (Table 6). Wage labor, in bothagriculture and non-farm business, also accrues primarily tothe poor. In contrast, white collar jobs such as medicine,

Table 4. Employment shares by activity and size of locality. Source: Bangladesh (2003), Banco Central de Chile (1986, 2002), and Zambia (2003)

Country (year) Totallabor

Agriculture Totalnon-farm

Manufacture Commerce andtransport

Personal, financial,and community services

Construction andutilities mining

ISIC code 1 2–9 3 6 and 7 8 and 9 2, 4 and 5

Bangladesh, 2000

Rural 100 58 42 10 17 12 3Intermediate urban 100 16 84 27 28 23 6Chittagong and Dhaka 100 8 92 26 29 32 5

Chile, 1984

Rural 100 65 35 5 9 17 4Intermediate urban 100 7 93 14 29 41 9Santiago 100 1 99 20 26 46 7

Zambia, 2000

Rural 100 90 10 1 2 7 1Intermediate urban 100 22 78 7 31 30 10Lusaka 100 0 100 14 22 54 10

Table 5. Returns to labor in rural non-farm activities, Darfur Sudan 1993.Source: Ibrahim (1997, chap. 2)

Non-farm activity Incomeper day (Sudanese pounds)

Sector

Tabag making 10 ManufacturingCarpet making 21 ManufacturingPot making 23 ManufacturingTea selling 60 CommerceWater peddling 75 CommerceFood selling 80 CommerceShoe making 150 ManufacturingBlacksmithing 150 ServicesConstruction 180 Services

Table 6. Capital intensity and returns to labor in rural non-farm activities,Bangladesh 1980. Source: Hossain (1984) cited in Lanjouw and Lanjouw

(2001)

Industry Capital perworker (Tk)

Value addedper worker(Tk/day)

Share of femaleworkers (%)

Tailoring 4,982 27.5 20Dairy products 3,076 23.4 10Gur (sugar) making 711 20.0 0Carpentry 3,009 19.9 4Jewelry 1,283 18.7 2Blacksmithy 760 15.8 2Handloom weaving 1,594 15.1 38Oil pressing 1,006 12.6 43Pottery 799 11.8 47Paddy husking 303 7.4 56Bamboo products 313 5.2 49Mat making 465 5.2 63Fishing nets 265 4.8 63Coir rope 145 4.1 64


teaching, accounting, and administration figure most promi-nently among higher income households (Table 7).

Given the low capital requirements and the small scale ofmany rural non-farm businesses, poor households dominatelarge segments of the rural non-farm economy. For this rea-son, many policy makers view the rural non-farm economy(RNFE) as a potentially important contributor to poverty

reduction. Others, however, fear that an abundance oflabor-intensive, low-return rural non-farm activities may sig-nal distress diversification and an absence of more productiveopportunities, given that low capital frequently translates intolow-productivity and low returns to labor (Table 6; Islam,1987; Shand, 1986).

Because of the differing equity impact of its various compo-nents and because of the differing composition of rural non-farm activity across settings, the overall impact of non-farmearnings on rural income distribution remains mixed (Table8). In some instances, aggregate non-farm earnings improveequity across household groups. In other cases, they exacer-bate inequality. And in some settings, the relationship betweenhousehold welfare and non-farm income shares proves

U-shaped. Empirically, no consistent pattern emerges. 2

4. DYNAMICS

(a) Shifting motors of rural economic growth

The present structure of the rural non-farm economy resultsfrom an ongoing economic transformation that has proceededfor many generations, though at varying speeds in differentlocations. At an aggregate level, structural transformation isa widely observed process by which productivity growthaccompanies a shifting sectoral composition of economicactivity. As productivity and incomes rise, households diver-sify their consumption out of foods, devoting an increasingshare of their earnings to the purchase of non-farm goodsand services. As a result, the share of agriculture in total na-tional output declines, and transfers of capital and labor outof agriculture help to fuel a corresponding rise in manufactur-ing and services. Since many of the resource flows from agri-culture to the secondary and tertiary sectors of the economytransit functionally and spatially via the rural non-farm econ-omy, an understanding of the forces that drive change in theRNFE becomes central to understanding the processes thatdrive overall economic growth.

Humans settle in rural areas where core economic activitiesenable them to earn a satisfactory living. In most developingcountry settings, agriculture forms the core of that rural eco-nomic base. Consequently, the dispersion of fertile soils, water,grazing land, and infrastructure largely govern the spatialdistribution of population across rural regions of the developingworld. Though less important in the aggregate, many ruralregions also contain natural resources—minerals, timber, or

Table 7. Income sources by quintile in rural India, 1999. Source: Lanjouw and Shariff (2004)

Quintile Agriculture Non-farm income Otherincome

Totalincome (%)

Real per capita

income (rupees)Cultivationagriculture

Agriculturewage labor

Wage labor Self-employment Regular employment Total non-farm

Lowest 38.2 28.2 15.8 11.4 4.4 31.6 2 100 1,146Q2 38 21.3 14.7 16.8 7 38.5 2.3 100 2,113Q3 45.2 13.4 10.1 16.3 11.7 38.1 3.2 100 3,141Q4 50.1 7.5 6.1 14.6 18.6 39.3 3.2 100 4,712Highest 64.5 2.1 2 7.9 21.1 30.9 2.5 100 11,226Total 54.9 8 5.9 11.5 17.1 34.4 2.7 100 4,468

Table 8. Mixed equity impact of rural non-farm income. Sources: Reardon et al. (1998), Lanjouw (1999), Adams (2002), and Lanjouw and Shariff (2004)

Quintile Rural non-farm income as share of total income

Equity enhancing Neutral Inequitable

Egypt, 1997 (%) Pakistan, 1989 (%) India, 1999 (%) Ethiopia, 1990 (%) Ecuador, 1995 (%) Viet Nam, 1997 (%)

Poorest 59 75 32 32 22 402nd 52 63 39 – 37 423rd 51 36 38 30 37 504th 53 33 39 – 46 60Highest 38 21 31 31 64 82


exotic natural settings—which sustain production of exportableraw materials, processed goods, and tourism services. Adminis-trative or entrepot trading centers can also serve as economicbase activities along transport corridors that transit remoterural locations. Overall, the economic base of a rural region in-cludes those core productive activities that remain competitivein external markets. Typically, these base activities producetradable outputs which face highly elastic demand and forwhich their rural location confers a comparative advantage.Total economic activity in the region, in turn, depends on thesize of the economic base and the demand it generates for localproviders of goods and services.

Agriculture, as the largest employer in rural areas, the larg-est income generator, and the largest purveyor of raw materi-als, clearly plays a predominant early role in influencing thesize and structure of the RNFE. Thus the process of ruraldevelopment normally begins with a countryside populatedby primarily agricultural households producing for themselvesmost of the farm and non-farm goods and services they re-quire. Trade and commerce remain marginal, given the subsis-tence orientation of agriculture, the prevailing low-input farmtechnologies, and the limited transport and communicationsinfrastructure in rural areas. In zones of rapidly growing agri-cultural productivity, the composition and patterns of growthobserved in the RNFE differ markedly from those in stagnantrural settings. For this reason, the following discussion distin-guishes between processes and outcomes operating in thesetwo very different environments.

(b) The “pull” scenario in dynamic agricultural regions

Where new agricultural technologies and modern farm in-puts become available, they lead to agricultural surpluses insome commodities and increased opportunities for trade. Inthese settings, a growing agriculture stimulates growth of theRNFE through a number of key linkages. Rising labor pro-ductivity on the farm increases per capita food supplies and re-leases farm family workers to undertake non-farm activities.For this reason, green revolution India has seen agricultural

labor fall from 75% to 65% of rural labor force in the first25 years following the release of green revolution rice andwheat varieties (Hazell & Haggblade, 1991). Equally impor-tant, increases in farm incomes, together with high rural sav-ings rates, make capital available for investment in non-farmactivities. These savings rates have reached up to 25–35% inmany areas of green revolution Asia (Bell, Hazell, & Slade,1982; Hazell & Ramasamy, 1991).

Increasingly productive modern agriculture requires inputsand services such as seeds, fertilizer, credit, pumps, farmmachinery, marketing, and processing of farm produce which,in turn, create a growing demand for non-farm firms that canprovide these services. Farm households, as their incomesgrow, increase their expenditure share on non-food items, 3

thereby accelerating demand for non-farm goods and servicessuch as housing, clothing, schooling, health, prepared foods,visits to town, to the cinema, and to the tea shops, all of whichdramatically increase demand for rural transport services. Tomeet this growing demand, rural households increasinglydiversify into production of rural non-farm goods and services(Table 9).

Some non-farm activities, initially undertaken part-time byfarm households for self-consumption, spin off as separatefull-time commercial activities. Others, particularly labor-intensive household manufacturing of baskets, pottery, androof thatching, die out, displaced by the import of cheapplastic pails, iron pots, and corrugated roofing. 4 For this rea-son, household manufacturing typically shrinks over time(Table 10).

The demise of low-productivity household manufacturingexplains, in part, why employment in services and commercefrequently grows faster than in manufacturing. Evidence from13 countries for which 10- to 20-year time-series data are avail-able suggest that rural manufacturing employment typicallygrows at only 1% per year while employment in rural tradeand services grows over three times as fast (Haggblade, Hazell,& Reardon, 2007, Table 4.4). Changes in consumer spendinglikewise contribute to faster growth in services and com-merce. Consumption data indicate that, as incomes rise, rural

Table 9. Growing rural non-farm income shares. Source: Oshima (1984),Hazell and Haggblade (1991), Poapongsakorn (1994, chap. 6), Hayami

(1998), Fan, Hazell, and Thorat (1999), Huang (1999), and Chadha(2003)

Rural non-farm income share of

Farm householdincome (%)

Ruralincome (%)

National income (%)

China

1978-80 17 – 41985 25 – 71990 26 – 101995 37 – 261997 39 – 28

India

1968 – 26 –1980 – 36 –2000 – 46 –

Japan

1950 22 – –1960 42 – –1970 63 – –1981 80 – –1987 84 – –

South Korea

1971 18 – –1981 33 – –1991 46 – –

Taiwan

1970 45 – –1975 47 – –1980 65 – –1987 78 – –

Thailand

1976 35 – –1986 46 – –

Table 10. Dynamics of rural non-farm employment growth, Bangladesh1990–2003. Source: World Bank (2004)

Sector Rural non-farm employment growtha

Permanent Household Total

Manufacturing 10.4 �0.9 0.6Services 3.4 5.0 4.5Total 4.5 2.9 3.3

a Average annual growth rate from 1990 to 2003.


households increase spending on services such as education,health, transport, cinemas, prepared foods, and transport fas-ter than they do on local manufactured goods. 5

As towns grow, they attract more workers from the ruralhinterland, leading to a rise in migration and even rural-to-ur-ban commuting. As a result, the share of agriculture in the to-tal workforce begins to decline, even though absolute levels ofagricultural output and employment may continue to grow forsome time. Over time, agriculture becomes progressively lessimportant as the economic motor for the regional economy.After several decades of agriculture-led growth, for example,India’s urban economic growth has stimulated corridors ofrural non-farm development along major highways and trans-port routes (Bhalla, 1981, 1997). Eventually, agriculture be-

comes a relatively minor economic activity in some ruralregions as well as in many national economies.

The composition of rural non-farm activity changes percep-tibly over time in these buoyant agricultural settings. Increasesin real wages raise the opportunity cost of labor, thereby mak-ing low-return non-farm activities uneconomic. This leads tothe demise of many low-return craft and household manufac-turing activities and to the growth of higher-return non-farmactivities such as mechanical milling, transport, commerce,personal, health, and educational services (Table 11). Growingagricultural incomes attract labor into more productive, high-er-return rural non-farm services.

In this setting, poor households benefit both directly andindirectly. Growing consumer demand translates directly intobusiness opportunities for self-employment, particularly inrural commerce and trade. Indirectly, rising rural wage ratesfor unskilled labor clearly benefit the rural poor.

(c) The “push” scenario in stagnant agricultural zones

In regions without a dynamic economic base, patterns ofgrowth in the rural non-farm economy unfold very differently.Sluggish income growth in agriculture leads to anemic con-sumer demand, limited agroprocessing and agricultural inputrequirements, and stagnant wages. Taken together, these ten-dencies stymie both entrepreneurial and wage-earning oppor-tunities in the rural non-farm economy.

Where population growth continues unabated for manygenerations, land availability diminishes and, ultimately, inthe absence of careful land management, so does soil fertility.Without technological advance in agriculture, labor produc-tivity and per capita farm production fall. In such settings,growing landlessness pushes labor force increments into non-farm activity by default (Table 12). Falling agricultural laborproductivity, low opportunity cost of labor, and declininghousehold purchasing power induce diversification into low-return, labor-intensive non-farm activities such as basket mak-ing, gathering, pottery, weaving, embroidery, and mat mak-ing. Specialized non-farm enterprises and householdsemerge, not to exploit potential productivity gains but becauseof an absence of opportunities in agriculture and a shortage ofboth rural savings and investible capital.

Declining economic conditions likewise motivate labormigration in search of more favorable opportunities elsewhere.Thus, in these agriculturally backward regions, migrationserves an important role as a regional safety valve as house-holds seek farm and non-farm employment opportunities inmore distant regions.

Rural residents seek out exchange opportunities in these set-tings, too, but mainly as they try to identify market outlets fortheir labor-intensive cottage industry goods. Rural towns,rather than accelerating agricultural advance, become evacua-tion points for rural labor and for labor-intensive rural non-farm exports. A stagnant, low-input agriculture generates littledemand for inputs or high-value repair, processing, and per-sonal services. Many describe this bleak downward spiral asa process of “agricultural involution.”

A study comparing stagnant and rapidly growing agricul-tural regions in India summarizes the contrasting situationsaptly.

“The difference between the faster growing agricultural areas and oth-ers is primarily in terms of the productivity and income levels in ruralindustries . . . Slowly growing agriculture not only fails to introduceany structural changes in rural industries . . . but also tends to keepthose engaged in rural industries at a subsistence level of productivityand income” (Papola, 1987, p. 104).

Table 12. Distinguishing good news from bad: contrasting sources of ruralnon-farm employment growth in a stylized Asian rice-growing economy.

Source: Haggblade and Liedholm (1991)

The greenrevolution(pull) (%)

The sponge(push) (%)

Initial shock Improvedagriculturaltechnology,a

labor using

Populationgrowthb

Resulting changes

Rural non-farm employment 1.9 1.9

Total rural employment 6.6 2.1Rural wage rate 6.6 �3.9Non-farm income 1.1 �4.7Total real per capita income 7.4 �4.4

Note: Using a semi-input-output model, with stylized data drawn from theMuda River region of Malaysia, these results simulate the following im-pact:a Labor-using technical change in agriculture that increases foodgrainoutput by 80% and is adopted by 50% of farmers.b Population growth rate of 6% over 3 years, just sufficient to generate anequivalent increase in rural non-farm employment.

Table 11. Labor market influences on the size and composition of rural non-farm activity, Bangladesh 1982. Source: Hossain (1988), pp. 95, 120

Income per hourin agriculturally underdeveloped

regions (Tk/h)

Percent by which agriculturally developed regions exceed underdeveloped areasa

Income per hourb (%) Employment (h/week) (%) Income per household (%)

Agriculture 5.1 29 8 40Non-agriculture

Services 11.4 4 30 35Cottage industry 4.4 90 �81 �63Wage laborc 2.8 6 �41 �38Trade 2.3 195 �28 113Total non-agriculture 4.4 59 �29 12

a Hossain distinguishes agriculturally “developed” and “underdeveloped” regions using a number of criteria: access to irrigation, use of modern ricevarieties, and fertilizer consumption, among others. In the agriculturally developed regions, modern varieties cover 60% of cropped area compared withonly 5% in underdeveloped areas.b Calculated from Hossain (1988), Tables 48 and 64.c Non-farm wage labor includes earth hauling, construction, transport, and “other” employment.


For poor households, falling wage rates and sluggish con-sumer demand result in diminished opportunities for bothbusiness start-up and wage earnings. In these settings, pros-pects for the rural non-farm economy remain bleak.

(d) Urbanization and migration

Although the prosperity of rural regions and their rural non-farm economies typically depends on agricultural performanceduring the early stages of economic growth, this link graduallyweakens over time as agriculture’s share in national economiesdeclines. Rapid urbanization and globalization have openedup new market opportunities for rural non-farm producersof tradable goods and services and for rural workers to mi-grate and remit. Where conditions permit, these opportunitiescan stimulate regional economic growth, in some instancesbenefiting backward regions with poor agricultural potentialand in others enhancing opportunities in already rapidly grow-ing rural economies.

Rising urbanization and national economic growth, to-gether with improved transport and communication networks,provide important economic linkages between urban and rural

areas, opening up new opportunities for rural households(Tacoli, 1998; Tacoli & Satterthwaite, 2003). Evidence fromIndia, for example, suggests that rapid rural non-farm growthis occurring along transport corridors linked to major urbancenters, largely independent of their agricultural base (Bhalla,1997). Similarly, in Southeast Asia and in China, high popula-tion density and low transport costs have led to rapid growthin urban-to-rural subcontracting for labor-intensive manufac-tures destined for international export markets (Otsuka, 2007,chap. 10).

The importance of migration and remittance incomeproves highly context specific, varying both locationallyand over time. While crises and civil strife may have reducedmigration opportunities in Africa during recent decades, size-able opportunities have emerged elsewhere (de Haan & Rog-aly, 2002). Most startling has been the growth in China,where migration has increased rapidly since the late 1980swhen government relaxed generation-long controls on popu-lation movement (Zhao, 1999). Between 1999 and 2003, thenumber of Chinese migrants doubled, from 52 to 98 million.China’s 2000 census suggests overall migration rates of 8.5%of the workforce, with roughly 30% heading to local town-ships, 30% to other counties in the same province, and40% representing movement across provinces (Du, Park, &Wang, 2005). Rural-to-urban commuting has also becomesignificant in some densely settled Asian and Latin Americansettings. A recent study from India, of six villages in AndhraPradesh and Madhya Pradesh, found that 10% of ruralhouseholds have at least one member commuting daily towork in nearby towns and cities, while a further 47% deploymembers in temporary migration (Deshingkar, 2004). Similarstudies from Chile and Peru document sharp increases indaily commuting from rural areas to nearby towns, particu-larly during the 1990s (Berdegue, Ramırez, Reardon, &German, 2001; Escobal, 2004).

Empirical evidence suggests that migrant remittances mayserve to increase rural investment, finance schooling, houseconstruction, and agricultural inputs in some locations (Francis& Hoddinott, 1993; Marenya, Oluoch-Kosura, Place, &Barrett, 2003). Less beneficial are the impacts on migrantworker health and on family social cohesion.

(e) Liberalization and globalization

Beginning in the 1990s, widespread economic liberalizationhas opened up the rural non-farm economy as never be-fore—to new opportunities and to new threats. Liberalization,


by reducing direct government involvement in production andmarketing, has opened up new market opportunities for theprivate sector, particularly in agricultural processing, inputsupply, and trade. Relaxed controls on foreign exchange andinvestment have unleashed a flood of foreign direct investmentinto Latin America, Asia, and Africa. As a result, large export-ers, agribusiness firms, and supermarket chains increasinglypenetrate rural economies of the developing world, alteringthe scale and structure of rural supply chains as they do (Rear-don, Timmer, Barrett, & Berdegue, 2003; Reardon, Stamoulis,& Pingali, 2007; Reardon & Timmer, 2005).

This rapidly changing environment opens up opportunitiesfor some rural suppliers to access new markets. In Bangladeshduring the second half of the 1980s, liberalization of the im-port of small diesel engines, as part of agricultural policy re-forms intended to boost irrigated rice production,simultaneously but quite unexpectedly launched a veritablerevolution in two major rural non-farm activities. After thecropping season, millers harnessed the new diesel engines topower 30,000 seasonal hammer mills, transforming the struc-ture of rice milling and dramatically increasing the competi-tion in rice markets. Later, during the rainy season, metalsmiths and boat makers adapted the engines to power thou-sands of river boats, converting these classic dhows fromcheap-but-slow to cheap-but-rapid inland water transport(Haggblade, Reardon, & Hyman, 2007; Jansen, Dolman, Jer-ve, & Rahman, 1989). Similarly in Thailand following WorldWar II, retired OSS operative Jim Thompson began workingwith silk weavers in Northeast Thailand, improving produc-tion technology and quality in rearing, reeling, and weaving.In the process, he helped to develop an international reputa-tion for the unique quality and design of Thai silk. The result-ing growth in silk exports transformed opportunities for over60,000 village women, enabling them to access internationalmarkets and roughly quadruple returns to labor in the process(Haggblade & Ritchie, 1992).

But liberalization and globalization expose other rural busi-nesses to new threats, as quantity requirements and qualitystandards impose new ways of doing business that risk exclud-ing undercapitalized rural enterprises on which the rural pooroften rely. In Latin America, where supermarkets now ac-count for over 60% of food retailing, the risks of this increas-ing market concentration have been most pronounced(Reardon, Timmer, Barrett, & Berdegue, 2003). Available evi-dence suggests that rapid concentration has triggered thebankruptcy of thousands of small firms in recent decades.Over 60,000 small food retailers closed their doors in Argen-tina from 1984 to 1993 (Gutman, 2002), while over 5000 smallfood retailers ceased operations in Chile during 1991–95 (Faig-uenbaum, Berdegue, & Reardon, 2002). Although many ofthese bankruptcies affected urban traders, emerging evidencesuggests that small rural traders and the wholesale marketsthey serve likewise risk being displaced by larger, specializedwholesalers. The leading supermarket chains in Guatemalaand Nicaragua, for example, have shifted away from ruralsmall-scale brokers for procuring fresh produce. Instead theynow procure through several large, specialized wholesalers(Balsevich, Schuetz, & Perez, 2006; Flores & Reardon, 2006;Hernandez, Reardon, & Berdegue, 2006). The same trendhas emerged in the beef sector in Nicaragua and Costa Rica(Balsevich et al., 2006).

Even in Sub-Saharan Africa, South African supermarketchains have expanded aggressively northward following theadvent of majority rule in 1994 and the demise of economicsanctions that had previously prevented these investments.

Two major chains, Shoprite and Pick N Pay, have opened out-lets in cities and rural towns in Zambia, Malawi, Mozam-bique, and Uganda and are considering forays into WestAfrica. In each locality, their entry has altered product selec-tion and market share in favor of imported South Africanbrands at the expense of local farmers, processors, food sup-pliers, and retailers (Weatherspoon & Reardon, 2003). Evi-dence from Kenya suggests that local supermarkets therehave similarly tended to replace rural traders and wholesalemarkets with specialized large-scale suppliers. While tradi-tional brokers and wholesale markets supplied 70% of freshfruits and vegetables in 1997, managers project that share willfall to 10% by 2008 (Neven & Reardon, 2004).

Some categories of rural non-farm activity have thrived inthe past because of protection from outside competition byhigh transport costs, restrictive production policies (such as re-served handicraft industries in India) and trade policies(including barriers to cheap imported consumer goods), subsi-dized inputs and credit, and preferential access to key markets(as in town and village enterprises in China). Globalizationand market liberalization remove many of these barriers, effec-tively “deprotecting” the RNFE. The transition may provebrutally abrupt for many traditional small-scale manufactur-ing activities whose products cannot compete with higherquality, mass-produced goods. For this reason, the initialstages of deprotection can lead to significant job losses inthe RNFE, even though many of these may later be recoveredas new types of rural non-farm activity sprout up, as in Indiaduring the 1990s (Bhalla, 1997). Since poor households and fe-male-dominated activities predominate among the low-invest-ment, low-productivity rural non-farm activities, they tend toface the most difficult adjustment during this transition.

(f) Summing up

Agriculture has historically played an important role inexpanding the economic base of rural regions in the develop-ing world. In regions where agriculture has grown robustly,the RNFE has also typically enjoyed rapid growth. A largegrowth linkages literature suggests that each dollar of addi-tional value added in agriculture generates $0.6–$0.8 of addi-tional RNFE income in Asia and $0.3–$0.5 in Africa andLatin America (Haggblade, Hazell, & Dorosh, 2007). Regionswith poor agricultural potential have seen more limited pros-pects for rural non-farm growth, except in places where theavailability of other important rural tradables such as mining,logging, and entrepot trade offers an alternative economicplatform for sustaining regional growth.

In recent years, globalization, urbanization, and improvedinfrastructure have opened up new opportunities in many rur-al areas, thereby reducing their dependence on agriculture.These developments seemingly offer new prospects for stimu-lating rural economic growth and, perhaps, new pathwaysout of poverty.

But just how powerful are these new opportunities and towhat extent have they substituted for agricultural growth asthe main driver of the RNFE? Although empirical evidencehas begun to trickle in, these processes are still unfolding inmany settings. So, firm conclusions will require more timeand new field studies. Based on the fragmentary evidenceavailable to date, it appears that the new forces of globaliza-tion and urban-led rural transformation are proving mostpowerful in densely populated, rapidly growing countries likeIndia and China. Tantalizing new evidence from India likewisesuggests that the correlation between agricultural growth and


growth of non-farm income and employment has becomeweaker in many rural areas (Foster & Rosenzweig, 2004;Harris-White & Janakarajan, 1997). In these settings, whereurban congestion, soaring rents, and higher wages raise thecost of doing business in metropolitan centers, rural-to-urbancommuting, temporary migration, and urban-to-rural subcon-tracting become economic (Otsuka, 2007, chap. 10). Increasedrural–urban flows can benefit rural areas by providing injec-tions of remittance earnings as well as growing urban marketsfor a range of rurally produced tradables. Opportunities forthis sort of urban-led rural growth appear more limited inpoorer, less densely populated, and more slowly growing eco-nomics, such as much of Africa. Controlling for national con-text, regions with better infrastructure and market access seemmore likely to benefit from urbanization and globalization,and these are often the better off regions to begin with. As aresult, rural areas with good infrastructure and market accessand which are located in successfully growing countries ap-pear likely to gain most from urbanization and globalization,while remote backward regions in poor, slow growing coun-tries will gain the least (Reardon, Stamoulis, & Pingali, 2007).

Overall, the key to stimulating expansion in the rural non-farm economy lies in sparking growth in key regional trad-ables—either in agriculture, tourism, or mining—or in linkingrural areas to external engines of economic growth that offerprospects for expanding rural markets and employment. La-bor market interactions—between agriculture and non-farmbusinesses and between rural and urban areas—offer crucialconnective tissue for linking rural workers, and particularlythe rural poor, to these growth processes.

5. IMPLICATIONS FOR THE RURAL POOR

(a) Motivation

The poor seek out opportunities in the rural non-farm econ-omy, not only to raise income levels but also to stabilizehousehold income over time. As a result, non-farm earningsserve an important safety net function. They provide a toolfor stabilizing household income during drought years (Gor-don & Craig, 2001; Reardon et al., 1998). Evidence fromsemi-arid areas of India suggest that non-agricultural self-employment and labor market earnings not only became amore important source of income on average during the arid1980s but also contributed to lower income variability (Walker& Ryan, 1990).

Given weak rural credit markets in many locations andlumpy seasonal cash flows in agriculture, non-farm earningsfrequently serve as an important source of liquidity forfinancing the purchase of modern hybrid seeds, fertilizer,and other inputs necessary for increasing farm productivity(Marenya et al., 2003; Oseni, 2007). A recent study from rur-al Nigeria finds that rural households with non-farm earn-ings roughly double their purchases of cash inputs inagriculture compared to households without non-farm earn-ings (Oseni, 2007). Across Africa, farmers typically dependmore heavily on non-farm income than farm input creditfor financing farm inputs (Reardon, Crawford, & Kelly,1994). Because non-farm income helps to finance on-farminvestments as well as purchased inputs, non-farm earningsprove important in helping poor households attain foodsecurity, both directly by helping the poor to buy foodand indirectly by financing the purchase of farm inputs nec-essary to increase food production.

(b) Constraints

Yet the paucity of their human, financial, and physical as-sets handicaps poor households in their efforts to participatein growing segments of the rural non-farm economy. As aresult, they often remain confined to the low-productivity,low-growth market segments which offer few pathways outof poverty, simply a means of bare survival. Meanwhile, richerand more educated households dominate the most lucrativenon-farm niches.

Gender, caste, and social status likewise restrict access bythe poor to the most lucrative non-farm activities in somesettings. In the same way that child-rearing obligationsmay limit women’s mobility and force them into home-based, highly labor-intensive pursuits such as weaving, silkrearing, and basketry, caste and social restrictions may forcespecific poor household groups into traditionally reserved,low-productivity rural non-farm activities. In India, these in-clude pottery, weaving, and tanning, among many others.Evidence from a range of settings demonstrates a correlationbetween asset poverty, ethnic minorities, and gender(Lanjouw, 2007; Lanjouw & Shariff, 2004). Discrimination,a weak asset base, and restrictions on geographic and occu-pational mobility all conspire to limit access by key disad-vantaged social groups to the most remunerative ruralnon-farm activities.

(c) Participation as entrepreneurs in growing supply chains

Poor households can potentially participate in an expandingrural non-farm economy in one of two ways: as entrepreneursin growing non-farm supply chains, or as employees in grow-ing segments of the labor market.

Typically, self-employment accounts for about 25% of ruralnon-farm employment. Indeed, in many settings, a majority ofnon-farm businesses are one-person firms, a proprietor withno employees. But management difficulties, weak credit mar-kets, and highly seasonal business activities lead to heavy turn-over, with high enterprise birth rates and correspondingly highclosure rates. Available evidence suggests that new firm start-ups account for about 20% of rural non-farm enterprises,while roughly 50% fail within the first three years (Liedholm,2007). Rural entrepreneurs constantly confront challengesand change.

Rapid change in many rural supply chains compounds thesemanagement challenges, requiring that rural firms and farmsswitch marketing channels, change how they do business, orinvest in equipment and organizational arrangements that en-able them to access growing market niches. Given disparitiesin economic power and access to information, these changesfrequently require collective action by alliances of small busi-nesses or brokering by pro-poor advocacy groups. Many dif-ferent models exist for attempting to facilitate market accessby small players in growing market niches. 6

(d) Participation as employees in growing segments of the labormarket

Not every poor person is a natural entrepreneur. Many lackthe information, mobility, technical skills, financial capital,and personality traits necessary to effectively manage a busi-ness. For them, the fastest route to prosperity lies in the labormarket. In buoyant regional economies, increasing wage ratesand labor demand fuel significant opportunities for even un-skilled labor (Table 11). Many of these opportunities emergein agricultural labor markets which often involve seasonal


rural-to-rural farm labor migration. Non-farm wage employ-ment also become available in rural construction businesses,processing mills, and assembly market networks. In some set-tings, government and private sector offer significant employ-ment opportunities for both unskilled and professionalworkers (Adams, 2002; Adams & He, 1995).

In some situations, non-farm employment opportunities re-quire commuting or temporary migration to rural towns andurban centers. Given the typical wage premiums prevailingin towns and cities, commuting or temporary migration fea-tures in many household diversification strategies. In poorprovinces of rural China, for example, roughly 25% of house-holds report at least one migrant member (Du et al., 2005). Toexploit available opportunities in regional labor markets, poorhouseholds benefit from improved information and transportflows; removal of social, economic, and administrative barri-ers; and investments in their human capital that permit themto expand labor earnings, the key resource poor householdspossess in abundance.

(e) Can RNFE growth slash poverty?

Empirical evidence on the link between RNFE growth andpoverty reduction requires careful sifting given the complexinter-relationships among agriculture, rural non-farm busi-nesses, and the national economy. Strong correlations betweengrowing rural non-farm income and falling rural poverty, as inChina since the 1980s (Ravallion & Chen, 2004), do not neces-sarily imply causality. Nor do they rule out the possibility thatindependent third factors, such as agricultural technology, maybe driving both (Lanjouw, 2007).

Available household panel data sets, which offer the bestmeans of isolating these effects, remain sparse. One such studyfrom rural Viet Nam concludes that growth of the rural non-farm economy is unlikely to serve as a primary motor of pov-erty reduction for the bulk of Viet Nam’s poor (Van de Walle& Cratty, 2003). In contrast, a village-level panel study fromIndia suggests that rural-based, export-oriented manufactur-ing may contribute significantly to rural poverty reduction(Foster & Rosenzweig, 2004). A study of migrant remittancesusing household panel data from poor regions of China findsthat having a migrant increases rural household per capita in-come by 9–13%. Yet the study concludes that the overall im-pact of migrant remittances on poverty remains modest,simply because most poor people do not migrate. The authorsestimate that migrant remittances decrease rural poverty head-count rates in poor regions of China by 1%, from 15.4% to14.4% (Du et al., 2005). A more general review of the migra-tion literature concludes similarly that, “Research on the ef-fects of migration on areas of origin is relatively scarce, butgenerally it shows that, at the macro-level, remittances con-tribute relatively little, and out-migration usually does notradically transform poor areas” (de Haan & Rogaly, 2002,p. 5).

However, overall rural non-farm earnings may reduce pov-erty significantly in specific settings. A time-series econometricstudy from China suggests that overall rural non-farm earn-ings, which accounted for 36% of rural income in 1997, havelikely proven highly significant in reducing rural poverty.Looking at different allocations of public expenditure, thestudy finds that government spending on education and ruralinfrastructure generates significant reductions in rural poverty,an impact they attribute largely to the stimulus these invest-ments provide to rural non-farm employment (Fan, Zhang,& Zhang, 2002). 7 Clearly, results differ across studies andacross settings.

Summing up, a recent review of available evidence con-cludes that the direct impact of rural non-farm earnings onrural poverty rates “proves muted in practice” but that non-farm activities nonetheless perform an important safety netrole, preventing families from falling further into poverty dur-ing crises. Moreover, the study suggests that the indirect effectsof rural non-farm employment, through labor market tighten-ing and rising rural wage rates, are “possibly very significant”for the rural poor (Lanjouw, 2007, pp. 78–79).

6. POLICY IMPLICATIONS

Policy makers hold high hopes that rural non-farm growthcan offer a pathway out of poverty for a large segment ofthe rural poor. As we have seen, prospects for RNFE growthprove brightest in well-connected rural regions in countrieswith rapidly growing agricultural and national economies. Incontrast, prospects are least favorable in remote backward re-gions, especially those embedded in poor, slow growing agri-cultural and national economies. The growing forces ofurbanization and globalization seem more likely to reinforcethan to change these patterns of development. As a result, itis unrealistic to expect that RNFE growth can solve povertyproblems in many backward regions of the developing world.However, it can play a significant role in regions with a strongrural economic base or with good access to growing urbaneconomies.

Given the enormous diversity observed across rural regionsand within the RNFE itself, opportunities, constraints, andappropriate policies will clearly differ across settings. Whilegeneral guidelines cannot substitute for detailed understandingof a specific rural non-farm setting, several broad policy guide-lines do emerge from this review.

Available evidence suggests that the rural non-farm econ-omy can significantly expand economic opportunities for therural poor if two conditions hold. First, the RNFE must itselfbe growing robustly. But rising non-farm employment will notsuffice. In stagnant rural regions, where a moribund agricul-ture offers little stimulus or employment growth, a surge ofgrowing population may simply push additional labor intoincreasingly low-return non-farm activities, as new entrants di-vide a fixed pie into increasingly smaller pieces. Rather, pov-erty-reducing rural non-farm growth requires an aggregateincrease in rural non-farm income coupled with growing in-come per worker. Typically, this requires investments in theproductive capacity and productivity of rural tradables suchas agriculture, tourism, or natural resource-based activitiesin order to ensure their competitiveness in external markets.Alternatively, where low-cost rural labor and low transporta-tion costs coincide, rural households can sometimes competein urban or export markets through commuting, short-termmigration, or urban-to-rural subcontracting arrangements.From a policy perspective, accelerating output and productiv-ity growth in the rural economic base will require investmentsin agricultural technology, rural education, communications,transportation, and electrification. Together with a favorablepolicy environment, these investments encourage rural non-farm business development as well as short-term commutingand migration strategies, both of which serve to increase ruralnon-farm incomes and investment.

But a growing rural non-farm economy does not guaranteeaccess by the poor. Wealthy households, well endowed withfinancial, human, and political capital, often prove betterequipped to take advantage of growth in the high-productivitysegments of the rural non-farm economy, both as entrepreneurs


and as wage employees. Meanwhile poor households, left totheir own devices, risk remaining relegated to slow-movingbackwaters of the rural non-farm economy. Migration oppor-tunities likewise remain bifurcated, with highly educated house-holds more apt to land lucrative positions in towns. Thus,policy makers cannot assume that an expanding rural non-farmeconomy will translate automatically into pro-poor growth.

This leads to the second requirement for pro-poor ruralnon-farm growth: access by the poor to growing non-farmmarket niches. In order for non-farm earnings to offer a path-

way out of poverty, rural households and policy makers mayneed to invest in rural education and health in order to im-prove the existing human capital stock of the poor. At thesame time, policy makers will need to remove existing eco-nomic and social barriers that limit entry by the poor intolucrative non-farm professions. Fluid labor markets, withgood transportation and communication systems connectingrural households to regional and urban labor markets, willprovide a key bridge linking the rural poor to growing oppor-tunities in the non-farm economy.

NOTES

1. See, for example, Anderson and Leiserson (1978) and Chuta andLiedholm (1979).

2. For an overview of the equity question, see reviews by Chuta andLiedholm (1979), Haggblade, Hazell, and Brown (1989), Bagachwa andStewart (1992), Reardon (1997), Reardon et al. (1998), Reardon, Taylor,Stamoulis, Lanjouw, and Balisacan (2000), Barrett et al. (2001), Lanjouwand Lanjouw (2001), and Lanjouw (2007).

3. See, for example, Mellor and Lele (1973), King and Byerlee (1978),Hazell and Roell (1983), and Hossain (2004).

4. Hymer and Resnick (1969) have dubbed these declining ruralmanufactures “Z-goods.” See Anderson (1982) for empirical evidence ofthis bifurcated transition.

5. See King and Byerlee (1978), Hazell and Roell (1983), Hazell andRamasamy (1986), and Hossain (2004).

6. See, for example, Boomgard, Davies, Haggblade, and Mead (1992),Montigard (1992), Chen (1996), Dowds and Hinojosa (1999), Bourgeoisand Herrera (2000), Lusby and Panliburton (2002), Kaplinsky and Morris(2002), Meyer-Stamer (2003), and Chitundu, Droppelmann, and Hagg-blade (2009).

7. Investment in agricultural R&D also proved highly effective inreducing rural poverty. These estimates project that 10,000 yuan spenton education pulls nine people out of poverty, while that same expenditureon agricultural R&D reduces poverty counts by 7. Spending on ruralroads, electricity and communications reduces poverty by 2–3 people foreach 10,000 yuan spent (Fan et al., 2002).

REFERENCES

Adams, R. H. Jr., (2002). Non-farm income, inequality and land in ruralEgypt. Economic Development and Cultural Change, 50(2), 339–363.

Adams, R. H., Jr., & He, J. J. (1995). Sources of income inequality andpoverty in rural Pakistan. Research report no.102. Washington, DC:International Food Policy Research Institute.

Anderson, D. (1982). Small industry in developing countries: A discussionof issues. World Development, 10(11), 913–948.

Anderson, D., & Leiserson, M. (1978). Rural enterprise and non-farmemployment. A world bank paper. Washington, DC: The World Bank.

Bagachwa, M. S. D., & Stewart, F. (1992). Rural industries and rurallinkages in Sub-Sharan Africa: A survey. In F. Stewart, S. Lall, & S.Wangwe (Eds.), Alternative development strategies in Sub-SaharanAfrica. New York: St. Martin’s Press.

Balsevich, F., Schuetz, P., & Perez, E. (2006). Cattle producers’ partici-pation in market channels in Central America: Supermarkets processors,and auctions. Staff paper 2006–01. Department of AgriculturalEconomics, Michigan State University.

Banco Central de Chile (1986). Indicadores Economicos y Sociales 1960–1985. Santiago: Banco Central de Chile, Direccion de Estudios.

Banco Central de Chile (2002). Chile social and economic indicators 1960–2000. Santiago: Banco Central de Chile, Direccion de Estudios.

Bangladesh, Bureau of Statistics (2003). Bangladesh population census,2001: National report. Dhaka: Bureau of Statistics.

Barrett, C. B., Reardon, T., & Webb, P. (2001). Non-farm incomediversification and household livelihood strategies in rural Africa:Concepts, dynamics and policy implications. Food Policy, 26(4),315–331.

Bell, C., Hazell, P., & Slade, R. (1982). Project evaluation in regionalperspective: A study of an irrigation project in Northwest Malaysia.Baltimore: Johns Hopkins University Press.

Berdegue, J., Ramırez, E., Reardon, T., & German, E. (2001). Rural non-farm employment and incomes in Chile. World Development, 29(3),411–425.

Bhalla, S. (1981). Islands of growth: A note on Haryana experience andsome possible implications. Economic and Political Weekly, (June 6),1022–1030.

Bhalla, S. (1997). The rise and fall of workforce diversification processes inrural India: A regional and sectoral analysis. Centre for economicstudies and planning, DSA working paper. New Delhi: JawaharlalNehru University.

Boomgard, J. J., Davies, S. P., Haggblade, S., & Mead, D. C. (1992). Asubsector approach to small enterprise promotion and research. WorldDevelopment, 20(2), 199–212.

Bourgeois, R., & Herrera, D. (2000). Actor-led change for efficient agrifoodsystems: Handbook of the participatory actor-based CADIAC approach.Bogor, Indonesia: CGRPT Centre: Regional Coordination Center forResearch and Development of Coarse Grains, Pulses, Roots and TuberCrops in the Humid Tropics of Asia and the Pacific.

Chadha, G. K. (2003). Rural non-farm sector in the Indian economy:Growth, challenges and future direction. Mimeo, Washington, DC:International Food Policy Research Institute.

Chen, M. A. (1996). Beyond credit: A subsector approach to promotingwomen’s enterprises. Ottawa, Canada: Aga Khan Foundation.

Chitundu, M., Droppelmann, K., & Haggblade, S. (2009). Intervening invalue chains: Lessons from Zambia’s task force on accelerationof cassava utilisation. Journal of Development Studies, 45(4),593–620.

Chuta, E., & Liedholm, C. (1979). Rural non-farm employment: A review ofthe state of the art. MSU rural development paper no. 4. East Lansing,Michigan: Department of Agricultural Economics, Michigan StateUniversity.

de Haan, A., & Rogaly, B. (2002). Migrant workers and their role in ruralchange. Journal of Development Studies, 38(5), 1–14.

Deshingkar, P. (2004). Rural–urban links in India: New policy challengesfor increasingly mobile populations. Paper presented at the World Bankrural week, March 2, 2004. <www.livelihoods.org/hot_topics/docs/UR_IndiaCaseStudy.doc>.

Dowds, C. M., & Hinojosa, J. (1999). An introductory guide to thesubsector approach to community and economic development. Washing-ton, DC: National Council of La Raza.

Du, Y., Park, A., & Wang, S. (2005). Migration and rural poverty inChina. Journal of Comparative Economics, 35(4), 688–709.

http://www.livelihoods.org/hot_topics/docs/UR_IndiaCaseStudy.doc

http://www.livelihoods.org/hot_topics/docs/UR_IndiaCaseStudy.doc


Ellis, F. (2000). Rural livelihoods and diversity in developing countries.Oxford: Oxford University Press.

Ellis, F., & Freeman, H. A. (2004). Rural livelihoods and povertyreduction strategies in four African countries. Journal of DevelopmentStudies, 40(4), 1–30.

Escobal, J. (2004). The role of public infrastructure in market developmentin rural Peru. Ph.D. dissertation. Development Economics Group,University of Wageningen.

Faiguenbaum, S., Berdegue, J. A., & Reardon, T. (2002). The rapid rise ofsupermarkets in Chile and its effects on the dairy, vegetable and beefchains. Development Policy Review, 20(4), 459–471.

Fan, S., Hazell, P., & Thorat, S. (1999). Linkages between Governmentspending, growth and poverty in rural India. Research report no. 110.Washington, DC: International Food Policy Research Institute.

Fan, S., Zhang, L., & Zhang, X. (2002). Growth, inequality and poverty inrural China: The role of public institutions. Research report no. 125.Washington, DC: International Food Policy Research Institute.

Ferreira, F. H. G., & Lanjouw, P. (2001). Rural non-farm activities andpoverty in the Brazilian Northeast. World Development, 29(3),509–528.

Fisher, T., Mahajan, V., & Singha, A. (1997). The forgotten sector: Non-farm employment and enterprises in rural India. London: IntermediateTechnology Publications.

Flores, L., & Reardon, T. (2006). Supermarkets new generation wholesalersand farmers’ organizations, contract farming and lettuce in Guatamala:Participation by and effects on smallholders. Staff paper 2006–2007. Department of Agricultural Economics, Michigan StateUniversity.

Foster, A. D., & Rosenzweig, M. R. (2004). Agricultural productivitygrowth, rural economic diversity and economic reforms: India, 1970–2000. Economic Development and Cultural Change, 52(3), 509–542.

Francis, E., & Hoddinott, J. (1993). Migration and differentiation inWestern Kenya: A tale of two sub-locations. Journal of DevelopmentStudies, 30(1), 115–145.

Gordon, A., & Craig, C. (2001). Rural nonfarm activities and povertyalleviation in Sub-Saharan Africa. Policy series no. 14. University ofGreenwich, Natural Resources Institute.

Gutman, G. E. (2002). Impacts of the rapid rise of supermarkets on thedairy products supply chains in Argentina. Development Policy Review,20(4), 409–427.

Habtu, Y. (1997). Farmers without land: The return of landlessness torural Ethiopia. In D. F. Bryceson, & V. Jamal (Eds.), Farewell tofarms: Deagrarianisation and employment in Africa. Hampshire,England: Ashgate.

Haggblade, S., & Ritchie, N. (1992). Opportunities for intervention inThailand’s silk subsector. GEMINI working paper no. 27. Bethesda,Maryland: Development Alternatives, Inc.

Haggblade, S., Hazell, P., & Brown, J. (1989). Farm-non-farm linkages inrural Sub-Saharan Africa: Empirical evidence and policy implications.World Development, 17(8), 1173–1201.

Haggblade, S., Hazell, P. B. R., & Dorosh, P. A. (2007). Sectoral growthlinkages between agriculture and the rural non-farm economy. In S.Haggblade, P. Hazell, & T. Reardon (Eds.), Transforming the ruralnonfarm economy: Opportunities and threats in the developing world(pp. 141–182). Baltimore: Johns Hopkins University Press.

Haggblade, S., Hazell, P. B. R., & Reardon, T. (2007). Transforming therural non-farm economy. Baltimore: Johns Hopkins University Press.

Haggblade, S., & Liedholm, C. (1991). Agriculture, rural labor marketsand the evolution of the rural non-farm economy. In G. H. Peters, &B. S. Stanton (Eds.), Sustainable agricultural development: The role ofinternational cooperation (pp. 542–557). London: Dartmouth Publish-ing Co.

Haggblade, S., Reardon, T., & Hyman, E. (2007). Technology as a motorof change in the rural nonfarm economy. In S. Haggblade, P. Hazell, &T. Reardon (Eds.), Transforming the rural nonfarm economy: Oppor-tunities and threats in the developing world (pp. 322–351). Baltimore:Johns Hopkins University Press.

Harris-White, B., & Janakarajan, S. (1997). From green revolution torural industrial revolution in South India. Economic and PoliticalWeekly, XXXII(25), 1469–1477.

Hayami, Y. (Ed.) (1998). Toward the rural-based development of commerceand industry: Selected experiences from East Asia. Washington, DC:World Bank.

Hazell, P., & Ramasamy, C. (1986). Household expenditure patterns andthe growth of the non-farm economy. (Mimeo) Paper presented at the

IFPRI/TNAU workshop on growth linkages, Oatacamund, India,February 14–16, 1986.

Hazell, P. B. R., & Roell, A. (1983). Rural growth linkages: House-hold expenditure patterns in Malaysia and Nigeria. Research reportno. 41. Washington, DC: International Food Policy ResearchInstitute.

Hazell, P. B. R., & Haggblade, S. (1991). Rural–urban growth linkages inIndia. Indian Journal of Agricultural Economics, 46(4), 515–529.

Hazell, P., & Haggblade, S. (1993). Farm-nonfarm growth linkages andthe welfare of the poor. In M. Lipton, & J. Van der Gaag (Eds.),Including the poor (pp. 190–204). Washington, DC: The WorldBank.

Hazell, P., & Ramasamy, P. (1991). Green revolution reconsidered.Baltimore: Johns Hopkins University Press.

Hernandez, R., Reardon, T., & Berdegue, J. (2006). Tomato farmerparticipation in supermarket market channels in Guatemala: Determi-nants and technology and income effects. Staff paper 2006–04. EastLansing, Michigan: Department of Agricultural, Food, and ResourceEconomics, Michigan State University.

Hossain, M. (1984). Productivity and profitability in Bangladesh ruralindustries. [Special issue on Rural Industrialization in Bangladesh].Bangladesh Development Studies, 2, 127–162.

Hossain, M. (1988). Nature and impact of the green revolution inBangladesh. Research report no. 67. Washington, DC: InternationalFood Policy Research Institute.

Hossain, M. (2004). Rural non-farm economy in Bangladesh: A view fromhousehold surveys. Center for policy dialogue (CPD) occasional paperseries no. 40. Dhaka: Center for Policy Dialogue.

Huang, J. (1999). Agricultural diversification and rural industrializationin China. In Paper presented at the symposium on critical food policyissues in Indochina in the emerging globalization context. Sponsoredby IFPRI and the Ministry of Agriculture and Rural Development,Hanoi.

Hymer, S., & Resnick, S. (1969). A model of an agrarian economy withnonagricultural activities. American Economic Review, 59(4), 493–506.

Ibrahim, H. B. (1997). Coping with famine and poverty: The dynamics ofnon-agricultural rural employment in Darfur, Sudan. In D. F.Bryceson, & V. Jamil (Eds.), Farewell to farms: Deagrarianisationand employment in Africa. Hampshire, England: Ashgate.

Islam, R. (1987). Rural industrialisation and employment in Asia. NewDelhi: ILO.

Jansen, E. G., Dolman, A. J., Jerve, A. M., & Rahman, N. (1989). Thecountry boats of Bangladesh: Social and economic development anddecision-making in inland water transport. Dhaka: University PressLimited.

Kaplinsky, R., & Morris, M. (2002). A handbook for value chain research.Ottawa: International Development Research Center (IDRC).

King, R. P., & Byerlee, D. (1978). Factor intensities and locationallinkages of rural consumption patterns in rural Sierra Leone. AmericanJournal of Agricultural Economics, 60(May), 197–206.

Lanjouw, P. (1999). Rural nonagricultural employment and poverty inEcuador. Economic Development and Cultural Change, 48(1), 91–122.

Lanjouw, P. (2001). Non-farm employment and poverty in rural ElSalvador. World Development, 29(3), 529–547.

Lanjouw, P. (2007). Does the rural non-farm economy contribute topoverty reduction?. In S. Haggblade, P. Hazell, & T. Reardon (Eds.),Transforming the rural nonfarm economy: Opportunities and threats inthe developing world (pp. 55–82). Baltimore: Johns Hopkins UniversityPress.

Lanjouw, J. O., & Lanjouw, P. (2001). The rural non-farm sector: Issuesand evidence from developing countries. Agricultural Economics, 26(1),1–23.

Lanjouw, P., & Shariff, A. (2004). Rural non-farm employment in India:Access, incomes and poverty impact. Economic and Political Weekly,39(40).

Liedholm, C. (2007). Enterprise dynamics in the rural nonfarm economy.In S. Haggblade, P. Hazell, & T. Reardon (Eds.), Transforming therural nonfarm economy: Opportunities and threats in the developingworld (pp. 99–114). Baltimore: Johns Hopkins University Press.

Livingstone, I. (1997). Rural industries in Africa: Hope and hype. In D. F.Bryceson, & V. Jamil (Eds.), Farewell to farms: Deagrarianisation andemployment in Africa. Hampshire, England: Ashgate.

Lusby, F., & Panliburton, H. (2002). Subsector/business service approachto program design, USAID. Washington, DC: Office of MicroenterpriseDevelopment, USAID.


Marenya, P. P., Oluoch-Kosura, W., Place, F., & Barrett, C. B. (2003).Education, non-farm income and farm investment in land-scarce WesternKenya. Basis brief no. 14.

Mellor, J., & Lele, U. J. (1973). Growth linkages of the new food graintechnologies. Indian Journal of Agricultural Economics, 18(1), 35–55.

Meyer-Stamer, J. (2003). PACA: Participatory appraisal of competitiveadvantage. <http://www.paca-online.de>.

Montigard, J. (1992). L’analyse des filieres agro-alimentaires. Economies etSocietes, Cahiers de l’ISMEA, 26(6), 59–70.

Neven, D., & Reardon, T. (2004). The rise of Kenyan supermarkets andevolution of their horticulture product procurement systems: Implica-tions for agricultural diversification and smallholder market accessprograms. Development Policy Review, 22(6), 669–699.

Oseni, G. (2007). Agricultural production, human capital and rural non-farmactivities: Evidence from Nigeria. PhD. dissertation. Washington, DC:American University.

Oshima, H. T. (1984). The significance of off-farm employment and incomesin post-war East Asian growth. Asian development bank economic staffpaper no. 21. Manila: Asian Development Bank.

Otsuka, K. (2007). The rural industrial transition in East Asia: Influencesand implications. In S. Haggblade, P. Hazell, & T. Reardon (Eds.),Transforming the rural nonfarm economy: Opportunities and threats inthe developing world (pp. 216–236). Baltimore: Johns HopkinsUniversity Press.

Papola, T. S. (1987). Rural industrialisation and agricultural growth: Acase study on India. In R. T. Shand (Ed.). Off-farm employment in thedevelopment of rural Asia (Vol. 1). Canberra: Australian NationalUniversity.

Poapongsakorn, N. (1994). Transformations in the Thai rural labormarket. In B. Koppel, J. Hawkins, & W. James (Eds.), Development ordeterioration? Work in rural Asia. Boulder, Colorado: Lynne RiennerPublishers.

Ravallion, M., & Chen, S. (2004). China’s (uneven) progress againstpoverty. Mimeo: DECRG, The World Bank.

Reardon, T. (1997). Using evidence of household income diversification toinform study of the rural non-farm labor market in Africa. WorldDevelopment, 25(5), 735–747.

Reardon, T., & Timmer, C. P. (2005). Transformation of markets foragricultural output in developing countries since 1950: How hasthinking changes? In R. Evenson, & P. Pingail (Eds.), Handbook ofagricultural economics (Vol. 3A). Amsterdam: North Holland Press.

Reardon, T., Crawford, E., & Kelly, V. (1994). Links between non-farmincome and farm investment in African households: Adding the capitalmarket perspective. American Journal of Agricultural Economics, 76(5),1172–1176.

Reardon, T., Stamolous, K., Balisacan, A., Cruz, M. E., Berdegue, J., &Banks, B. (1998). Rural non-farm income in developing countries. Thestate of food and agriculture 1998 (pp. 283–356). Rome: Food andAgricultural Organization of the United Nations.

Reardon, T., Stamoulis, K., & Pingali, P. (2007a). Rural nonfarmemployment in developing countries in an era of globalization.Agricultural Economics, 37(1), 173–183.

Reardon, T., Stamoulis, K., & Pingali, P. (2007b). Rural non-farmemployment in developing countries in an era of globalization. In K.Otsuka, & K. Kalirajan (Eds.), Contributions of agricultural economicsto critical policy issues. Malden, MA: Blackwell.

Reardon, T., Taylor, J. E., Stamoulis, K., Lanjouw, P., & Balisacan, A.(2000). Effects of non-farm employment on rural income inequality indeveloping countries: An investment perspective. Journal of Agricul-tural Economics, 51(2), 266–288.

Reardon, T., Timmer, C. P., Barrett, C., & Berdegue, J. (2003). The rise ofsupermarkets in Africa, Asia and Latin America. American Journal ofAgricultural Economics, 85(5), 1140–1146.

Shand, R. T. (1986). Off-farm employment in the development in rural Asia.Canberra: Australian National University.

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

Tacoli, C., & Satterthwaite, D. (2003). The urban part of rural develop-ment: The role of small and intermediate urban centres in rural andregional development and poverty reduction. Rural–urban interactionsand livelihood strategies working paper 9. London: InternationalInstitute for Environment and Development.

Van de Walle, D., & Cratty, D. (2003). Is the emerging nonfarm marketeconomy the route out of poverty in Vietnam? Policy research workingpaper no. 2950. Washington, DC: The World Bank.

Walker, T. S., & Ryan, J. G. (1990). Village and household economies inIndia’s semi-arid tropics. Baltimore, MD: Johns Hopkins UniversityPress.

Wanmali, S. (1983). Service centres in rural India: Policy, theory andpractice. Delhi: B.R. Publishing Co.

Weatherspoon, D. D., & Reardon, T. (2003). The rise of supermarkets inAfrica: Implications for agrifood systems and the rural poor. Devel-opment Policy Review, 21(3), 333–355.

World Bank (2004). Promoting the rural non-farm sector in Bangladesh(Vol. 2). Rural Development Unit, South Asia Region, Team led byForhad Shilpi (August). Washington, DC: World Bank.

Zambia, Republic of (2003). Zambia 2000 census of population andhousing: Main census report. Lusaka: Central Statistics Office.

Zhao, Y. (1999). Leaving the countryside: Rural-to-urban migrationdecisions in China. American Economic Review, 89(2), 281–286.

.sciencedirect.com
Available online at www
http://www.paca-online.de

the rural non-farm economy: prospects for growth and poverty reduction

Documents