afghanistan research and evaluation unit - · pdf filedeveloping world (2004); ... the...

Afghanistan Research and Evaluation Unit

Synthesis Paper Series

Urban Livelihoodsin Afghanistan

August 2006

Jo Beall and Stefan Schütte

Funding for this researchwas provided by theEuropean Commission (EC)

About the AuthorsJo Beall is Professor of Development Studies and Director of the Development StudiesInstitute (DESTIN) at the London School of Economics and Political Science (LSE). She is adevelopment policy and management specialist, with expertise on urban development andurban governance. Her other interests include gender, social policy and internationaldevelopment, social exclusion and local responses to crisis and conflict. She is currentlyresearching conflict cities and the role of cities in state making in contexts of state weaknessand fragility. Professor Beall has conducted extensive research in Southern Africa and Southand West Asia and has undertaken research consultancies for a range of internationaldevelopment agencies. She is author of Cities and Development (forthcoming 2007) andFunding Local Governance (2005); co-author of Urban Governance, Voice and Poverty in theDeveloping World (2004); Uniting a Divided City: Governance and Social Exclusion inJohannesburg (2002) and Sustainability (2002); editor of A City for All (1997); and co-editorof Fragile Democracy: State and Society in Democratic South Africa, Journal of SouthernAfrican Studies 31(4) December, 2005 and Policy Arena on the Discourse of Terrorism,Security and Development, Journal of International Development 18(1) 2006.

Stefan Schütte has a doctoral degree from the South Asia Institute of the University ofHeidelberg, Germany. He is a social geographer who has worked for the last eight years inIndia and Afghanistan. His work includes extensive research into social and economic changein a tea-growing village in South India, when farmers shifted from subsistence to cash cropcultivation. Subsequently, Dr. Schütte moved on to urban research and worked for threeyears on the livelihoods of marginalised urban communities in the North Indian city ofVaranasi. Since early 2004, he has worked with the Afghanistan Research and EvaluationUnit managing the urban livelihoods project.

About the Afghanistan Research and Evaluation Unit(AREU)The Afghanistan Research and Evaluation Unit (AREU) is an independent research organisationthat conducts and facilitates action-oriented research and learning intended to inform andinfluence policy and practice. AREU also actively promotes a culture of research and learningby strengthening analytical capacity in Afghanistan and by creating opportunities for analysisand debate. Fundamental to AREU’s vision is that its work should improve Afghan lives. AREUwas established by the assistance community working in Afghanistan and has a board ofdirectors with representation from donors, the United Nations, multilateral organisationsand non-governmental organisations (NGOs).

Current funding for AREU is provided by the European Commission (EC), the United NationsAssistance Mission in Afghanistan (UNAMA), the United Nations High Commissioner for Refugees(UNHCR), Stichting Vluchteling and the governments of the United Kingdom, Canada, Denmark(DANIDA), Switzerland, Norway and Sweden. Funding for this study was provided by theEuropean Commission.

Acknowledgements

This report summarises 12 months of intensive fieldwork in 120 poor urban households inKabul, Herat and Jalalabad. We would like to thank all the families who participated andwhose experiences form the very basis of this research synthesis. We are grateful to themfor their remarkable cooperation and patience throughout the study period. They welcomedus warmly during our regular visits, which intruded upon their precious time and privacy.

A study of this type and scale would not have been possible without our team of researchers,who established trust and rapport with the members of our study households: Parvin andHayatullah in Herat; Dr Sharaf and Basira in Jalalabad; and Leena and Anil in Kabul. Theydid a magnificent job. Special thanks to Mamiko Saito, who contributed significantly to thesuccess of the project.

The entire project benefited tremendously from the guidance and management supportprovided by Dr Paula Kantor, whose substantive input to the content and direction of thestudy is invaluable.

We would also like to acknowledge Aftab Opel for helping us design the methodology in theinitial stages of the project and for his overall input. The authors wish to thank Daniel Esserfor his substantive input throughout the process of producing this report.

Finally, we want to thank Jolyon Leslie, Paul Fishstein and Pushpa Pathak for their veryuseful comments on the drafts of this report and, last but not least, Fe Kagahastian for hereditorial contribution.

Jo Beall and Stefan Schütte, Kabul, August 2006

Table of Contents

Glossary .................................................................................................. i

Acronyms ................................................................................................ ii

Executive Summary .................................................................................... 1

1. Introduction .................................................................................... 7

2. The Political Economy of Afghan Urban Livelihoods ..................................... 9

2.1 Background ............................................................................ 9

2.2 Sustainable urban livelihoods: Framing the analysis .......................... 11

3. Methodology .................................................................................. 17

4. Land, Services and Social Infrastructure ................................................. 21

4.1 Land, housing and security of tenure............................................ 22

4.2 Basic services: Water supply, sanitation and electricity ..................... 26

4.3 Health and education .............................................................. 28

5. Urban Labour Markets and Household Consumption.................................... 35

5.1 Urban labour markets and income-earning opportunities .................... 35

5.2 Household structure and the division of labour ................................ 42

5.3 Coping with risk: Consumption, credit and debt............................... 46

6. Asset Building through Collective Action and Representation ........................ 51

6.1 Neighbourhood and family networks............................................. 51

6.2 Social assets and collective action ............................................... 52

6.3 Social assets and representation ................................................. 54

7. Urban Livelihoods and Implications for Policy........................................... 59

7.1 Key findings .......................................................................... 59

7.2 Implications for national policy................................................... 60

7.3 Implications for urban policy and city planning................................ 62

7.4 Conclusion............................................................................ 63

References..............................................................................................65

GlossaryAfghani (or Afs) official Afghan currency

ashar communal, reciprocal work (typically around life cycle events)

gharibkar work activities carried out by the poor

ograyi form of credit in which goods are loaned to seller and paid for afterfinal sale

mozarebat partnership agreement in which one person provides money or assetsand other time and labour, from which profits are shared

wakil-e gozar neighbourhood representative

wasita relations to powerful people

Zor Abad informal settlement; literally, “place taken by force”

Afghanistan Research and Evaluation Uniti


AcronymsAKDN Aga Khan Development Network

AREU Afghanistan Research and Evaluation Unit

DFID Department for International Development (UK)

EC European Commission

GoA Government of Afghanistan

I-ANDS Interim Afghanistan National Development Strategy

IDP Internally Displaced Person

MoUD Ministry of Urban Development

NDF National Development Framework

NSP National Solidarity Programme

SLF Sustainable Livelihoods Framework

UNDP United Nations Development Programme

Afghanistan Research and Evaluation Unit ii

Urban Livelihoods in Afghanistan

The cities of Afghanistan are growing fast.The pace of this growth exceeds the planningand management capabilities of the alreadyoverwhelmed central government and ofunder-resourced municipalities, making themunable and, at times, unwilling to effectivelywork on reducing levels of urban poverty andvulnerability. As a result, poor urbanhouseholds are excluded from the most basicservices and are forced to live day-to-day onmeagre, unreliable income flows in oftenhealth-threatening environments.

Against this background of unprecedentedand largely unplanned urbanisation in thecountry, the Afghanistan Research andEvaluation Unit (AREU) carried out researchin five Afghan cities to explore the groundrealities of urban poverty and vulnerability.By understanding the diverse livelihoodstrategies poor and vulnerable householdspursue, and how and why such strategieschange over time, this AREU study hopes toinform policy formulation that builds on thecapabilities, needs and priorities ofmarginalised urban populations.

The report draws on empirical data fromindividual studies conducted across five urbanlocations. Three case studies — in Kabul,Herat and Jalalabad — involved longitudinalresearch among 40 households over a 12-month period. In addition, short-term citycase studies were conducted in Mazar-i-Sharifand Pul-i-Khumri over a two to three-weekperiod in each location. In the first case, thelife paths of poor urban households wereobserved and tracked over an entire year.Economic changes and shifts in the livelihoodtrajectories of households were documentedand the movements of households trackedalong a continuum of livelihood security andvulnerability. In the Mazar-i-Sharif and Pul-i-Khumri cases, research focused on specificlivelihood situations in poor urban settlementsin northern Afghanistan, capturing thedynamics in intermediate-sized cities. This

Executive Summary

synthesis report presents key findings and acomparative analysis of the results fromacross these study sites.

Key findings

Access to land, services and socialinfrastructure

• Afghanistan’s poor urban population havelittle or no access to basic services andsocial infrastructure. This is a result oflimited resources, combined with theauthorities’ unwillingness and lack ofcapacity to serve effectively. Exclusionfrom basic services adversely affects thecapacity of the urban poor to earnadequate income and acquire thenecessary human assets to have qualityof life. While lack of services has longbeen recognized as a major problem inurban Afghanistan, investments in safewater supply, sanitation, and shelterprogrammes have been largelyinsufficient.

• Part of the problem lies in the fact thatmunicipalities and other governmentinstitutions tend to ignore informalsettlements that lack legal status,although they make up a significant shareof the population in Afghanistan’s majorcities. The unwillingness to develop apro-poor land policy that acknowledgesinformal settlements, and the absenceof a coherent vision that includes thesesettlements in urban planning lead to thesocial exclusion of citizens. Their rightto the city is neglected by a malfunction-ing planning system that fails to recognizethe urgency in releasing sufficient landfor legal occupation.

• The minimal to non-existent provision ofthe most basic human needs to thoseresiding in informal settlements forcesfamilies and communities to develop

Afghanistan Research and Evaluation Unit 1


burdensome and frequently health-threatening coping strategies. This is mostobvious when it comes to water supply.Many urban households have no accessto clean water and are driven to spendtheir already limited income on privatewater supply. Families who cannot affordto do so end up relying on polluted surfacesources or travelling long distances to thenearest public pump or well.

• Polluted water, along with the absenceof sanitation systems, contribute tochronic health problems. Illnesses thatcould be prevented are instead madeworse by the absence of a health facilitynearby and the high cost of privatehealthcare. This prevents many remotecommunities from seeking treatment forchronic illnesses or emergency care, muchless preventive check-ups. Exclusion fromsuch basic services makes families highlysusceptible to diseases and healthproblems, compromising their ability toearn a living.

• Access to education is problematic formost poor urban households acrossAfghanistan. Children are required eitherto work for an income, or assist inperforming household chores. For manychildren living in remote communities,attending school is nearly impossible withthe distant location of learning facilities.

Urban labour markets and householdconsumption

• Informal employment, which lacksprotection and income security, is themain livelihood source for the greatmajority of Afghanistan’s urbanpopulation. While it provides poorhouseholds with an immediate source ofincome, families end up being morevulnerable with irregular, low-paying jobsthat force them to deploy women andchildren — at times, to exploitative andhazardous types of work — for additionalincome. Moreover, households are often

forced to be highly mobile to adapt tochanging labour demands across differentlocations and seasons. As a result, theyexperience constant variation in incomelevels, reducing their ability to plan forthe future.

• The two biggest types of employment forthe bottom strata of the urban labourforce are self-employment and casualwage labour, both of which arecharacterised by low and erratic incomesand high seasonality. Home-based work,as the third biggest type of employment,is largely the domain of women. It ischaracterised by even lower incomes.Dependency on home-based work variedacross study sites. For example, home-based income activities account for morethan a third of all paid work in Herat,whereas in Jalalabad, such work hardlytakes place.

• Household structure and composition arekey determinants of income levels:households able to mobilise male labourare better off than those unable to doso. There is a critical link between urbanvulnerability and health, particularly thatof male breadwinners. Their physicalwellbeing impacts their ability to securefood, which in turn influences the healthof the entire family.

• Insecurity of employment leads to incomeirregularity and chronic shortage of cashfor the urban poor, making it hard forthem to establish savings. Faced with thecomparatively high cost of living in Afghancities, a huge majority of householdsresort to borrowing money just to be ableto afford basic food consumption, letalone pursue investments in healthcareand housing. In all cities, except for Herat,average expenditures for basic necessitiesexceed average incomes for members ofthe study population. A great majority ofthe households are in constant debt as aconsequence.

Afghanistan Research and Evaluation Unit2


Household resources, social assets andrepresentation

• The asset base of the studied urbanhouseholds is not sufficiently diversifiedto lift them out of poverty in any lastingway. Their reliance on scant andinconsistent resources adds pressure ontheir coping mechanisms. This is the casefor families that live on paid labour, aswell as those relying on the goodwill andsupport of others.

• The most vulnerable households also shiftresidence regularly, both in pursuit ofwork and to avoid rising rent. This deprivesthem of a more permanent place to callhome and decreases their chances ofacquiring legal housing, which representsthe most important physical asset for theurban poor.

• Crisis and emergency situations are usuallycushioned by social relations, notablythat of family and relatives. They serveas a safety net, an indispensable socialasset for urban livelihood security.

• Informal social networks rarely reachbeyond neighbourhoods or next of kin andthere is very little evidence of efforts topromote or expand community relationsamong the urban poor in Afghanistan. Amajority of the studied householdsappeared to have strong socialconnections that are indispensable tomaintain ing exist ing l ivel ihoodopportunities. However, their networksessentially serve as a means to survival;the resources acquired through themhardly allow urban families to rise abovepoverty.

• In addition, such social networks tend tobe overused. Without basic materialresources in place, mutual assistance isdifficult to sustain and leaning consistentlyon social networks could overburden andcompromise relations. Informal socialprotection systems may quickly erode,especially for households with the most

precarious livelihood strategies, who tendto borrow more than they could pay.These families are left even morevulnerable: once they have exhaustedthe resources of relatives and friends,they are left without a safety net.

• There is little evidence of poor andvulnerable households being involved inlocal level decision-making processes.This relates, in part, to the control ofcommunity organisations held by maleelders, who do not always represent themore marginalised members. Thus, poorhouseholds bear the brunt of hugebacklogs in service delivery and areexcluded from urban planning.

• A responsive government is central tosuccessful representation. The city casestudies provided little evidence of cityor national government responsivenessto the needs and interests of the poor.It is clear that there is a lack of capacity,resources and political will among civilservants.

Recommendations

Stimulate broad-based labour intensivegrowth at the national and city levels

• Evaluate Afghanistan’s open marketapproach and consider the possibleeconomic and employment gains fromimport substitution and protectionism insome key industries, such as cement andcotton/textiles.

• Ensure that opportunities for good qualitywork are valued as much as the quantityof jobs created. Thus, a holistic strategysupporting urban employment creationmust include market analysis of demandfor higher value products and the labourand skill requirements associated withthem. This must be combined with thedevelopment of vocational training andformal education programmes to developa labour force with the appropriate skills.



• Develop a broad planning framework thatanticipates the continuing increase inurban populations, resulting from theinflux of returnees from neighbouringcountries, as well as from pull factorsdrawing rural residents to urban areas.

• Facilitate the participation of women inthe urban labour force under improvedconditions, where their work is valuedappropriately and they are given a realchoice about place of work. In addition,the burden of household work must bereduced by supporting community childcare initiatives and improving access tobasic services.

Redress the imbalance caused by thepredominantly rural focus of nationaldevelopment assistance in Afghanistan

• Design and implement nationwide urbanpoverty reduction programmes to improvethe quality of life of the urban poor at ascale approaching the reality of theproblem. These programmes mustrecognize the inter-linkages between themyriad problems underlying urban povertyand take an integrated versus sectoralapproach, linking job creation, skillsbuilding and urban service provision, aswell as improving urban governance.

• Streamline coordination and cooperationbe tween l i ne m in i s t r i e s andmunicipalities. Much of the backlog inservice provision and barriers to pro-poorurban development derive from theinability and unwillingness of governmentagencies to work together. Incentivesmust be developed to ensure effeciency.Successful coordination structures fromother countries could be studied andadapted in developing effective urbangovernance institutions in Afghanistan.

• Increase donor support for thedevelopment and implementation ofurban poverty reduction and capacitybuilding programmes within government

institutions. Supporting sustainable urbandevelopment requires medium to longterm financial commitments, not short-term funded pilots that are not broughtto scale. Donor support must becomprehensive to achieve successfulinstitutional reforms.

Increase the human capital of the urbanpoor through improved access to qualityhealthcare and education

• Extend affordable and proximate healthcare facilities into or close to poor urbanneighbourhoods, allowing low incomegroups to regularly access bothpreventative and curative care.

• Plans for the allocation of new schoolsmust be made in a way that distributesaccess equitably across urban locationsto promote school attendance.

• Efforts to provide proper school buildingsshould be accompanied by greater effortsto ensure quality teaching and teachingsupport.

Develop risk mitigating social protectionprogrammes, supporting a shift away fromreliance on the coping capacity andresourcefulness of the urban poor

• Develop and finance national socialprotection programmes to support themost vulnerable (families without ablebodied male workers, families dependenton disabled members for income, etc.)and to invest in programmes supportingasset diversification for the poor to assistthem out of poverty.

• Provide donor support for urban povertyreduction programmes that go beyondphysical upgrading and technical serviceprovision to address social insecurities.

• Build the political will and capacity ofnational and city level actors (bothgovernment and non-government) toimprove their abilities to address social



protection issues at the required scale inthe urban sector.

Negotiate a path between the reality ofburgeoning informality and the rigidity offormal master planning

• Build capacity among municipal leadersand staff to move them away from asolely technical planning approach linkedto a master plan. Help them accept theimportance of including citizens in theplanning process and to develop moreflexible tools reflecting the on-the-groundreality of all urban residents, particularlythe poor.

Recognize and legalise informal settlements

• Develop a national urban land policy inline with the aim of equitabledevelopment. Establish guidelines andcategories for municipalities that wouldfacilitate recognizing and legalisingexisting informal settlements. The policymust also consider how to accommodatethe minority living on untenable land.

• Provide security of tenure to residents inrecognized informal settlements, withrecommendations for approaches to landtitling coming from an analysis of bestpractices in other similar post-conflictand Islamic contexts.

Ensure universal coverage with basicservices (water, drainage and sanitation,electricity, access roads); movingincrementally toward this goal will require:

• The political will among state institutionsto serve the poor.

• Improved communication betweengovernment institutions involved in servicedelivery, including a detailing of rolesand responsibilities and establishing ameans of coordination.

• Improved communication and coordinationbetween government and non-governmentagencies involved in service provision,for better distribution of responsibilitiesand to share lessons learned and bestpractices.

• Community consultation, involving bothwomen and men, to devise appropriateservice standards and allocate operationand maintenance responsibilities.

• Technical creativity in devising ways todeliver services to some harder-to-reachareas, ensuring that the work of disparateagencies links up at the city level.

• Recognition of the employment creationpotential of service delivery andinvolvement of poor urban residents(women and men) as small scaleentrepreneurs.

Acknowledge the important contributionof informal economic actors to the city

• Vendors and street sellers and their rolein the urban market should beacknowledged, not as a nuisance, but asimportant contributors to the economiclife of the city. Recognition can comethrough some form of licensing,transparently developed to avoid chancesof unscrupulous rent-seeking.

• The needs of street vendors and theimportance of local markets must berecognized in city plans; they must beplanned for and not marginalised.

• Institute an inclusive consultation processto discuss how to regulate informaleconomic activity and to ensure that theneeds of informal actors and themunicipalities are represented. Thismeans ending harassment and rent-seeking and guaranteeing that vendingactivities do not impinge on the effectiveoperation of municipalities.



1 Government of Afghanistan (2004), Securing Afghanistan’s Future, Kabul: GoA.

All major cities in Afghanistan haveexperienced significant population growthduring the last decade. Conflicts, droughtand refugees returning home after the fallof the Taliban in 2001 have all contributedto the rapid pace of urbanisation unfoldingin major cities. This trend is ongoing and hasbeen accelerated by growth rates twice ashigh as in the rural areas of the country, withan estimated doubling of urban populationsin Afghanistan by the year 20151. Con-sequently, challenges for urban planning andmanagement are manifold. Governmentagencies, municipalities and the internationalaid community alike are compelled to thinkabout how to best address increased urbanpoverty and vulnerability occurring in thewake of such rapid growth. Despite thisunprecedented and largely unplannedurbanisation in the country, little is formallyknown and documented about the living

conditions of urban residents in Afghanistan,and the real life challenges families face asthey try to survive day-to-day.

It is against this background that theAfghanistan Research and Evaluation Unit(AREU) carried out research on urbanlivelihoods and vulnerability in five Afghancities. AREU recognizes the need to addressthis crucial knowledge gap and to develop adetailed understanding of: the diverselivelihood strategies carried out by poor andvulnerable urban populations; the contextsin which people pursue particular livelihoods;their ability to access services; and how andwhy their adopted livelihood strategies changeover time. This research has been undertakenwith the overall objective of informing thenational policy formulation process — toinfluence it in such a way that policy makingresponds more meaningfully to the needs and

1. Introduction



priorities of poor and vulnerable urbanfamilies and builds on their energy andcapabilities.

The research also aims to help actors involvedin international development assistance toimplement appropriate, diverse, effectiveand practical interventions that strengthenthe livelihoods of the urban poor. To achievethis objective, the research design developedindicators that explain why some householdsmanage to achieve longer term wellbeingand even build asset portfolios, while othersendure persistent poverty. In doing so, theresearch helps illustrate and explain theheterogeneous nature of urban poverty inAfghanistan.

This report draws on empirical data fromindividual studies conducted across five urbanlocations. As a synthesis report, it presentskey findings from across these study sites,as well as a comparative analysis of thelongitudinal research carried out in threelarge urban centres: Kabul, Herat, andJalalabad; and short-term research in two

additional sites: Mazar-i-Sharif and Pul-i-Khumri2. In the first case, the life paths ofpoor urban households were observed andtracked over an entire year. Economic changesand shifts in the livelihood trajectories ofhouseholds were documented and themovements of households tracked along acontinuum of livelihood security andvulnerability. The complementary short-termcase studies in Mazar-i-Sharif and Pul-i-Khumriwere undertaken with a focus on specificlivelihood situations in poor urbansettlements, in order to cover a major cityin northern Afghanistan and capture thedynamics in an intermediate-sized city.3

Building on the findings of these five individualcity case studies, this synthesis report providesanalytical commentary on the emergingthemes and trends that are critical forunderstanding and responding to urbanlivelihood insecurity across all cities. Thereport also offers readers local context andpeculiarities that distinguish each location,exploring both differences and similaritiesbetween study sites.

2 Schütte, S. (2006), Searching for Security: Urban Livelihoods in Kabul, Kabul: AREU; Schütte, S. (2006), Gaining SomeGround: Urban Livelihoods in Jalalabad, Kabul: AREU; Schütte, S. (2006), Poverty amid Prosperity: Urban Livelihoods inHerat, Kabul: AREU.

3 Schütte, S. (2006), Poor, Poorer, Poorest: Urban Livelihoods and Vulnerability in Mazar-i-Sharif, Kabul: AREU; and SchütteS. (2006), Dwindling Industry, Growing Poverty: Urban Livelihoods in Pul-i-Khumri, Kabul: AREU.



2.1 Background

In 2001, Afghanistan emerged from 22 yearsof war to face increased economic, politicaland social challenges. In the wake of aprotracted drought (1999-2001) and a decadeof relentless political violence through yearsof civil war and Taliban oppression, thepredominantly rural economy was severelyweakened. Livelihoods were devastated bythe decimation of livestock and agriculturalproduction. Falling incomes, loss of assetsand the ravages of war saw over 5 millionpeople displaced as refugees in neighbouringcountries. As a recent World Bank reportsuggests, “Afghanistan was essentially leftout of the last 25 years of global development,with virtually no increase in per capita incomeduring this period and an average lifeexpectancy of only 43 years.”4 Indeed,Afghanistan remains one of the world’spoorest countries, with an average per capitaGDP of US$300 per annum (including opium).It has persistently poor social indicators anda gender gap only exceeded by Niger.5.........

From this disheartening base the economyhas recovered fairly well, although at adeclining rate.6 To a significant degree,economic growth has been associated withthe cessation of drought and war, as well asthe benefits of reconstruction, such as thebooming construction industry. Sustainingthe targeted nine percent growth rate willrequire higher agricultural yields, lessdependence on the poppy economy anddiversification of economic activities withthe growth of other sectors.

Cities play a crucial role in stimulating thenational economy through local economic

development and the facilitation of regionaland international trade. At present manyeconomic opportunities — and relatedlivelihood opportunities — are confined tothe informal economy. While it provides poorurban families with an immediate means oflivelihood, operating in this unregulatedinformal economy also means no job security,earning income from temporary work thatdo not always translate into sustainablelivelihoods. Moreover, when income isirregular, there is less capacity to affordbasic services, much less own a home or savemoney for the future.

For internally displaced persons (IDPs),resources are even more limited. As refugeesforced to live a transient life across rural-urban and regional and national boundaries,it is much more difficult for them to establishlinks with social networks. As a result, theylose an important safety net, given thataccess to employment, housing and loansdepends on knowing people with influenceand who are widely connected within acommunity. Opportunities that do arise as aresult of growing political stability andreconstruction do not usually extend to thosewho are unconnected and vulnerable.

IDPs and refugees, who returned after thecessation of hostilities in 2001, found littleto support a living in the countryside. Insearch of adequate income, many familiesmade their way to Afghanistan’s national andprovincial capitals. Even those who remainedin the rural areas diversified their livelihoodsby sending some household members to townsand cities. This intensified competition forurban livelihood opportunities. While accuratestatistics remain unavailable, it is generally

2. The Political Economy of Afghan Urban Livelihoods

4 World Bank (2004), Afghanistan: State Building, Sustaining Growth and Reducing Poverty, A Country Economic Report,Poverty Reduction and Economic Management Sector Unit, South Asia Region, Report No. 29551-AF, 9 Sept 2004, p. xvi.

5 UNDP (2003), Human Development Report: Millennium Development Goals: A Compact among Nations to end HumanPoverty, New York: UNDP.

6 GDP growth rates were 25-30 percent in 2002-03, 15-20 percent in 2003-04 and estimated at 8 percent in 2004-05. SeeUNDP (2004), Afghanistan National Human Development Report: Security with a Human Face. Kabul: UNDP; and GoA (2005),Afghanistan Statistical Yearbook 2005. Kabul: GoA.



accepted that Afghanistan’s urban populationhas been growing at an inexorable pace sincethe fall of the Taliban, adding to thepopulation increases experienced during theSoviet occupation and mujahidin period. Itis estimated that the urban population nowconstitutes between 23 to 30 percent of thetotal population of the country.7 It is notclear, however, whether these trends are atemporary consequence of the collapse ofrural livelihoods or a more permanentphenomenon, although there is evidence tosuggest that once established, households doconsolidate themselves in cities. Thisresearch, as a longitudinal study, contributesto a better understanding of these trendsand the characteristics of urban livelihoods.

Afghanistan’s political institutions andprocesses have been severely disrupted byyears of conflict. By the year 2000, the centralstate had virtually collapsed as a legitimatepolitical entity. It failed to maintain authorityand security and it remains weak in terms of

administrative capacity, unable to implementeffective policies and deliver public services,particularly to the poor. In theory, Afghanistanhas a highly centralised state but in reality,it has experienced a form of de factodecentralisation. In many cases, the relativeautonomy of regional authorities has beenbuilt around illegal seizure of local resourcesand revenue streams and is not alwaysaccompanied by local level public investment.“Warlordism” and the absence of the “ruleof law” have been fuelled by opiumproduction and the narcotics trade, whichthrive in lawlessness and insecurity. It hasbeen argued that “the opium economy is thelynchpin of a “vicious circle” of insecurity,weak government, powerful warlords anddrug money,”8 with the latter feeding intomilitary assets that support local and regionalfiefdoms.

Clearly, Afghanistan’s economic and politicalfuture requires a vigorous approach to nationalunity and central state building, a strategy

7 Beall, J. and D. Esser (2005), Shaping Urban Futures: Challenges to Governing and Managing Afghan Cities, AREU IssuesPaper Series, Kabul: AREU, March 2005, p. 7.

8 World Bank (2004), Afghanistan: State Building, Sustaining Growth and Reducing Poverty, p. xvi.



explicitly expressed in the Government’sNational Development Framework (NDF) andS e c u r i n g A f g h a n i s t a n ’ s F u t u r e :Accomplishments and the Strategic PathForward (SAF).9 A new Constitution andsuccessful presidential and parliamentaryelections have constituted important firststeps toward stability and nationalgovernance. As the political process continuesto evolve, one outstanding challenge is toreform the bureaucracy. While publicadministration is highly centralised on paper,it has weak links with and control overprovinces and municipalities, which, takentogether, have been largely ineffective todate in achieving local level economicdevelopment, governance and servicedelivery. There is a critical tension betweenthe national imperative to strengthen thecentre and the need at the city level toensure that effective municipal structuresand processes are developed and supported.At present, only a small proportion of thepopulation, mainly the urban and provincialelites, makes a decent living and has accessto local level services such as good health-care, education, safe water supply andsanitation. Access to urban labour markets,infrastructure and services must be extendedto the poor and vulnerable. This will involvebreaking down existing power structures andremoving the influence of vested interests— goals which may be formidable yet notimpossible to achieve.

2.2 Sustainable urban liveli-hoods: Framing the analysis

The term “livelihoods” has been used to referto assets, income-generating activities andthe access to both (mediated by institutionaland social relations). Together, theseelements determine an individual’s or a

household’s quality of life.10 The idea ofsustainable livelihoods became very influentialin the development community among thosewho recognized the need to move beyondmoney-metric concepts of poverty to includea more holistic and dynamic understandingof deprivation. Influenced by Amartya Sen’sanalysis of poverty and Robert Chambers’exploration of risk and vulnerability,development practitioners started looking atlivelihoods with a multidimensional approachto social disadvantage, paying attention topeople’s vulnerabilities, capabilities andendowments.11 Such an approach regardeda household’s livelihood sustainable when itallows the family to cope with economicstresses, recover from shocks and consistentlymaintain or enhance its financial capabilitiesand assets without depleting its naturalresource base.

A common feature across existing livelihoodsframeworks is the assessment of the strengthof people’s “capitals” or “assets”. One ofthe most widely adapted frameworks is theSustainable Livelihoods Framework (SLF),developed by the United Kingdom’sDepartment for International Development(DFID). The SLF identifies five types of capital:human, financial, physical, social andnatural.12 Human capital constitutes thelabour resources available to households andtheir quality. Financial capital refers to thesavings, credit, remittances and pensions,which allow households different incomestreams and livelihood options. Physicalcapital comprises basic infrastructure suchas housing, transport, water supply andcommunications, as well as the equipmentand tools by which people pursue theirlivelihoods. Natural capital refers to naturalresources useful to livelihoods such as land,water and public resources: rivers or common

9 GoA (2004), Securing Afghanistan’s Future: Accomplishments and the Strategic Path Forward (SAF), Kabul: GoA.10 Ellis, F. (2000), Rural Livelihoods and Diversity in Developing Countries, Oxford: Oxford University Press, p. 10.11 Sen, A.K. (1999), Development as Freedom, Oxford: Oxford University Press; Chambers, R. and G. Conway (1992), Sustainable

Rural Livelihoods: Practical Concepts for the 21st Century, IDS Discussion Paper No. 296, Brighton:IDS, University of Sussex.12 Carney, D. (1998), Implementing the Sustainable Livelihoods Approach, in D. Carney (ed.), Sustainable Rural Livelihoods:

What Contribution Can we Make? London: Department for International Development, pp. 3-23.



land. Social capital consists of relationshipsbased on trust and membership in a group orcommunity. Such social networks are reliedupon for employment, loans and other typesof assistance. All of these assets help familiesundertake household responsibilities and gainaccess to labour and more institutions ofsociety.

The various frameworks reviewed for thestudy also include analysis of vulnerability,the likely trends and shocks that couldinfluence livelihoods, as well as the structuresand processes that determine who gets accessto markets, shelter or decision-making arenas.The latter is very important to livelihoodanalysis in the Afghan context, where systemsand structures often operate on the basis ofwasita — knowing influential people, whocould provide access to livelihood assets andeven positions of power. An analysis of urbanlivelihoods in Afghanistan demonstrates thatreciprocal networks of support, whileindispensable as an informal social protection,can sometimes become an obstacle to poorand vulnerable groups accessing decisionmakers beyond their community. For example,a weak wakil-e gozar may fall under thecontrol of a commander and stand in the wayof other citizens, who are more active indeveloping political assets through directdealings with local officials, politicians orNGOs.

One criticism about the SustainableLivelihoods Framework is its inadequateemphasis on institutional factors andgovernance issues, falling short in addressingthe significance of politics and the role playedby voice and rights in determining livelihoodoutcomes. These critiques are particularlyimportant to understanding the dynamicsaround urban livelihoods and to adapting aframework that factors in social asymmetries,the complexities of the institutional context

and the prominent role played by politicsand power in urban settings.13

Perceptions of poverty in many urban areasare not only associated with materialdeprivation but also include the more psycho-social issues of respect and dignity, stigmaand exclusion. Also relevant is an assessmentof which forms of social relationships peopletend to align themselves with and how suchconnections relate to the immediate needsof their households, for instance, land andlabour. It is important to recognize that insome cases, poverty is not necessarilyexacerbated by exclusion from resources oremployment, but by power imbalances, whichinfluence how people gain access to resources.Individuals with less power in socialrelationships may have to make trade-offsbetween dependence on a source foremployment, land or credit needed forimmediate survival, and the ability to makeindependent choices that could lift them outof poverty in the medium to long term. Theybecome locked in negative social relations,which may provide security today but at thecost of long term growth prospects.14 Forexample, a family may obtain land throughconnections to a commander who grabbedit. This provides short-term security of tenurethrough the commander’s protection butcould then lead to future insecurity if therelation to the commander deteriorates, orland policy reforms are implemented,nullifying the commander’s authority overthe grabbed land.

In general, it has been a challenge to capturesuch dynamics in a livelihoods frameworkintended to address urban issues.Nevertheless, a strategic approach, includinga framework specific to a certain setting, isnecessary in tackling urban problems. Havinga framework to start with allows planners tolink micro-level analysis of how poor peoplecope and prosper, with how city institutions

13 Some authors suggest the inclusion of a sixth category: “political capital,” claiming that social capital does not capturethe complexity of urban social and institutional interactions. See Beall, J. (2005), “Urban Livelihoods” in C. Toumain andB. Wisner (eds), Towards A New Map of Africa, London: Earthscan Publications, pp. 97-112.

14 Wood, G. (2003), “Staying Secure, Staying Poor: The ‘Faustian Bargain’”, World Development 31(3): 455-71.



support or hinder opportunities for them toachieve quality of life. This linkage isparticularly important in analysing urbanlivelihoods in the Afghan context, whereformal and informal institutions largely hinderindividual and collective efforts to securelivelihoods. Part of the obstacle to povertyreduction is the lack of capacity and politicalwill among local government officials, whosepersonal interests often come before officialduties. This prevailing attitude may not onlylead to corrupt practices, it also deprivesthe poor of the services and security theyneed in order to have a chance at pursuinga better life.

In the Afghan context, a livelihoods frameworkcan structure an analysis of how poor urbandwellers navigate their way through a highlyimperfect institutional environment (state,market, NGOs/civil society, family); how they

negotiate within different levels andstructures of power to achieve the bestpossible outcome for their livelihoods. As itstands, this “best possible” outcome falls farshort of the national government’s goal ofachieving equitable development and pro-poor growth.

Figure 1 provides a framework adapted tourban settings, highlighting the key elementspertaining to a livelihoods perspective. Inrelation to the vulnerability context, thestudy findings clearly demonstrate thatseasonality is an inevitable feature of low-income employment in urban areas. Moreover,some of the households investigated in thisstudy combined seasonal work in cities withagricultural labour.

In addition, urban dwellers are subject toshocks and stresses experienced in cities.

Figure 1. A framework for understanding sustainable urban livelihoods15

15 Adapted from DFID’s Sustainable Livelihoods Guidance Sheets; and Rakodi, C. and T. Lloyd-Jones, eds (2002), UrbanLivelihoods, A People-Centred Approach to Reducing Poverty, London: Earthscan, p. 9.



Urban Infrastructure and Services andthe Role of Municipalities

Vulnerabilitycontext:

ShocksStressesSeasonsaffecting urbangroups.These may bedriven bydynamicsoperating atinternational,regional,national, city andlocal levels.

Policies,institutionsand processes:

At national,international,regional and citylevels. These includethe structure ofgovernment anddecentralisationprocesses; levels ofinterstate transfersand fiscal autonomy;and laws; illegality,culture and socialpractices; and the“rules of the game”in place at the citylevel.

URBAN

LIVELIHOODS

Urban Income-Earning Opportunitiesand the Urban Economy

Naturalassets

Physicalassets

Financialassets

Humanassets

Social andpolitical assets

Many towns and urban communities acrossAfghanistan have yet to recover from decadesof conflict. Inflationary pressures also makethe cost of living particularly difficult forpoor urban dwellers. In Kabul, for example,an increased demand for housing due to theinflux of foreigners, returnees and IDPs hasled to rent hikes and pressure on land in thelast several years. Confounding the problemwas the prevalence of land grabbing — aprofound shock to many families who losttheir properties. The winter months alsoprovide a regular source of stress for poorurban dwellers, who could barely afford thehigh costs related to fuel and ill health.

The policy and institutional environment inFigure 1, the box on the right, representsthe role of institutions in the livelihoods ofthe poor. It is here that power, inequalityand the terms of inclusion of the poor enterinto the analysis of urban livelihoods. Thus,it is a central element of the framework thatcan provide not only a realistic assessmentof the coping mechanisms of the urban poor,but also of the constrained environment inwhich many families struggle to eke out whatis often meagre living.

Figure 1 also highlights the role ofmunicipalities as an important area foranalysis. In Afghanistan, their role relatesprimarily to land, housing, and local economicdevelopment but there can be areas ofoverlap and tension between municipalitiesand line ministries, particularly in the capitalcity, Kabul. The responsibilities of localauthorities link to urban livelihoods in twoimportant ways. In the first place they helpdetermine how effectively people canundertake their day-to-day productive andreproductive activities. For example, cheapand reliable transport allows poor people inperipheral settlements to travel daily tocentral labour and commodity markets.Municipal functions are also critical in ensuringthat people not only survive but are able to

save and prosper in a vibrant local economicenvironment. Below is a sample of thediscussions clarifying the key role of localleadership vis-à-vis urban disadvantage:

Urban poverty is much influenced bywhat city municipal governments door do not do; also by what they canor cannot do. As an understanding ofpoverty widens — for instance toinclude poor quality and/or insecurehousing, inadequate services and lackof civil and political rights — so doesthe greater current or potential roleof local government to contribute topoverty reduction.16

It is clear that the urban poor and thelivelihoods they pursue are closely linkedwith municipal structures and patterns oflocal governance that largely determine theiraccess to or exclusion from necessaryresources. In the Afghan urban context, thesestructures and patterns are often inefficient,impeding the urban poor’s search forsustainable livelihoods. Such institutionaldeficiencies may be attributed to a numberof factors, including lack of political will,lack of capacity and resources, and a strugglebetween fulfilling personal interests and thestate’s responsibilities.

In general, cities are regarded as the enginesthat drive economic growth, wheregovernments and their partners are best ableto deliver infrastructure and services on scale.This is not always the case, however. Theassociation between urbanisation and risingper capita income does not always hold andeven when it does, not all urban dwellersbenefit from the growth. In fact, some cities,in their efforts to become or remaincompetitive, introduce automation intoindustries and production processes, pushingwages down and excluding unskilled labour.

In Afghanistan, as the case studies show,trends in urban livelihoods are dictated by

16 Satterthwaite, D. (2000), Editor’s Introduction: “Towards more Pro-poor Local Governments in Urban Areas”, Environmentand Urbanisation, 11(2), pp. 3-11.



the shape of the urban economy in specificcities, all of which are characterised by smalland stagnant labour markets or urban growthpoles that reinforce inequality more thanthey address problems of poverty. Thus, theability of Afghan cities to drive growth ispatchy, to say the least, and even morequestionable when viewed through the prismof public service delivery and employmentgeneration.

The findings of the research also illustratethat acquiring jobs and income-earningopportunities are a fundamental pre-occupation for Afghanistan’s urban poor. Thisis hardly surprising given their almost totaldependence on cash for survival. It must benoted, however, that pursuing income-earningactivities outside of the home goes hand inhand with care giving obligations within thehousehold. Productive work, after all, requiresgood physical health and the durability ofurban households depends on the wellbeingand education of younger and futuregenerations. Without a decent livingenvironment, it will be nearly impossible tostay healthy and productive, much less planfor the future. It is, therefore, important toexamine how the responsibilities for variousactivities within and outside the householdare distributed; and how such distributioninfluences livelihood opportunities andoutcomes and the general wellbeing ofindividual members and households as awhole.

A livelihoods perspective offers an integratedframework for understanding the relationshipbetween access, resources and power andits implications for asset building.17 Anyanalysis of livelihoods must incorporate an

assessment of intra-household dynamicsbecause nowhere is the complex relationshipbetween access, resources and power moreacutely observed than in the household. Theresearch on Afghan cities shows how familydynamics, in and of itself, can be considereda livelihoods asset. Indeed, family supportto income-earning activities can be crucialin explaining the survival or growth of oneurban household over another in a similarexternal environment.18 The research showshow the composition and structure of urbanfamilies (who is the head of the household,for example) influence a household’s abilityto mobilise labour and diversify strategies tomaximize income — at times to the detrimentof other members. This report, as well asthe individual city studies on which it isbased, is informed throughout by a genderperspective and sensitivity to householdstructure, headship and dynamics betweengenerations.

In poor countries, such as Afghanistan,providing any form of social protection isextremely difficult. Moreover, given thecountry’s history of political and economicturmoil, natural disasters and the ravages ofwar, the national response has been orientedtoward household survival and promotingindividual enterprise, rather than effortstoward redistribution, something that is beingreinforced by current state policy and advicefrom international development agencies.19

This government approach may very well bea recognition of the Afghan people’s“widespread personal and private initiative”;that they are a “resilient, creative,opportunity-seeking and entrepreneurialpeople.”20 One can argue that:

17 Bebbington, A. (1999), Capitals and Capabilities: A Framework for Analysing Peasant Viability, Rural Livelihoods and Poverty,World Development, 27(12), pp. 2021-2044.

18 Moser, C.O.N. (1996), Confronting Crisis, Environmentally Sustainable Development Studies and Monograph Series, No. 7,World Bank; and Moser, C.O.N. (1998), The Asset Vulnerability Framework, World Development.

19 Ministry of Rural Rehabilitation and Development (2003), Nationwide Risk and Vulnerability Assessment (NRVA), Kabul:MRRD; and World Bank (2004), Afghanistan: State Building, Sustaining Growth and Reducing Poverty, A Country EconomicReport, Poverty Reduction and Economic Management Unit, South Asia Region, Report No. 29551-AF.

20 World Bank (2005), Afghanistan: Poverty, Vulnerability and Social Protection: An Initial Assessment, Human DevelopmentUnit, South Asia Region, Report No. 29694-AF, p. i.



Poverty and vulnerability inAfghanistan, while extensive, havealways been and should continue tobe principally managed by the poorthemselves. As such, the challengefor future governments will be toharness and support the provencapabilities of the Afghan peopletoward collective national objectivesfor development and prosperity.21

The resilience and fortitude of the Afghansis not in question. However, the researchfindings show that it is important not to seerural or urban dwellers as eternally energeticand resourceful, and local communities asconstant repositories of social care. In manycases, poverty makes it nearly impossible forindividuals to survive, pushing them to relyon others in their households and socialnetworks to make ends meet. In such

conditions, it is unacceptable for the stateand other development agencies to temperor even withhold intervention based on theresilience and resourcefulness of poor Afghans.

This research also shows that when peopleare above the subsistence level — in urbanareas as much as anywhere else — socialsupport systems and networks that offermutual benefits continue to be operative andeffective. However, when conditions are direand shocks and stresses are relentless, thisreciprocal support could come under a lot ofstrain. Reciprocity entails expectations andobligations, often not equal in outcome.When households in a community networkcome under heavy economic stress, conflictsarise not only in relation to access toresources, but also around whose interestsare represented in appeals made to local andcity level decision making processes.

21 Ibid.



With rapid urbanisation unfolding inAfghanistan, it is necessary to have acomprehensive and contextual understandingof urban livelihoods. AREU has conducted anumber of earlier studies that provided auseful basis on which to build this moreextensive research.22 This study took adistinctive longitudinal approach to urbanlivelihoods in Afghanistan and was based onintensive fieldwork in three urban locations.This was supplemented by one-off fieldworkconducted in two additional cities in orderto check for regional differences within theresource and time constraints involved.

In the long-term study areas — Kabul, Herat,and Jalalabad23 — researchers worked witha permanent sample of 40 households in eachsite. They were visited monthly over a periodof one full year, allowing for in-depthinvestigation of their circumstances and anychanges over time. This fieldwork produceddetailed evidence of the multiple livelihoodand coping strategies used by different typesof households and the same households atdifferent times. The rationale for thisapproach was based on awareness that tounderstand urban poverty properly, researchmethods providing “snapshots” at a particularmoment in time were inadequate. Alongitudinal approach was consideredessential in order to understand why, in somany cases, poverty persists and how in somecases, people do manage to find ways out ofpoverty. By mapping livelihood trajectoriesover time and by observing and documentingprocesses and changes to which people areexposed, how they cope with them and howthey attempt to induce change themselves,this analysis provides a more nuancedunderstanding of the causal factors at work

in relation to urban poverty and vulnerability.

The selection of longitudinal study sites wasdriven by the aim to include, as much aspossible, Afghanistan’s major cities and tocapture the ethnic diversity of the country.There was also a concern to exploredifferences that may arise in terms of policies,institutions and processes that could affecturban livelihoods. Obviously, Kabul, as thecapital and as a city that has exploded insize in recent years, had to be included. Itwas also important to include Jalalabad inthe East as a Pashtun dominated border cityclose to Pakistan. Critical too was the inclusionof Herat in the West with its wide ethnicdiversity and close proximity to Iran, believedto have a strong influence on the city’slivelihood activities. Afghanistan has twoadditional major cities: Mazar-i-Sharif andKandahar. The former was included as ashort-term supplementary case study tocapture the dynamics at work in the largesturban centre in the north of the country. Theadverse security situation in Kandahar madeimplementation of this part of the researchplan impossible but it did allow the team toinclude, as a substitute, the medium-sizedcity of Pul-i-Khumri, which served as aninteresting case study because of its smallersize but comparatively large industrial base.

The initial identification of neighbourhoodsand households taking part in the three long-term study areas was accomplished by meansof focus group discussions in differentneighbourhoods, with local communitycouncils, as well as through collaborationwith NGOs working in the areas. In addition,researchers sought to include different typesof households in the sample: those with

22 Beall, J. and D. Esser (2005), Shaping Urban Futures: Challenges to Governing and Managing Afghan Cities, AREU IssuesPaper Series; Grace, J. (2003), 100 Households in Kabul: A Study of Winter Vulnerability, Kabul: AREU; Hunte, P. (2004), SomeNotes on the Livelihoods of the Urban Poor in Kabul, Afghanistan, Kabul: AREU; Schütte, S. (2004), Urban Vulnerability inAfghanistan: Case Studies from Three Cities, Kabul: AREU.23 For detailed city descriptions and their contextual peculiarities, see the individual case study reports by Schütte.

3. Methodology



diverse ethnicities, female-headed householdsand those headed by people with disability,returned refugees, or internally displacedpersons (IDPs). This way, a broader samplewas identified first, then gradually reducedto a select sampling of 40 households basedon the diversity of household types, as wellas pragmatic considerations, such as ahousehold commitment to long-termparticipation in the research. Regular visitswere made to households included in thefinal sample, with the first visit in February2005 and continuing with repeated monthlyvisits through January 2006.

Through intensive fieldwork, the followingaspects and issues were explored:

• Narrative household histories andhousehold activity profiles using semi-structured questionnaires;

• Inter- and intra-household relations, creditrelations, physical asset base and accessto basic services, migration activities,and urban–rural linkages, using a set ofstructured questionnaires implementedonce during the 12-month period;

• Economic changes through detailedmonthly monitoring of labour marketparticipation, income and consumptionfigures, credit and indebtedness and self-help and support activities among all 40households, with the support of astructured questionnaire; and

• Changes in activity and asset portfoliosand in livelihood conditions among a sub-sample of 10 selected households, throughdetailed monthly qualitative monitoringinterviews. These narrative interviews,guided by a checklist, were typically verytime-intensive, which is why it wasdecided to carry them out in a sub-sampleof households.

Due to time constraints, a differentmethodological approach was applied to theshort-term study sites of Mazar-i-Sharif and

Pul-i-Khumri. In these two sites and inconsultation with NGOs and localmunicipalities, poor urban settlements wereselected and group discussions held withrepresentatives of the local communities.These discussions addressed a variety ofthemes and provided a picture ofneighbourhoods, their histories and the statusof infrastructure and service delivery fromthe perspective of community members. Theresearch here also explored perceptions ofdifferences in the livelihood situations of themembers of these communities. This initialstep was then complemented byquestionnaire-guided interviews with families,bringing in a household perspective.

This rather ambitious research programmewas carried out without major problems,although in the long-term study sites, workingwith the same households over an extendedperiod with a high frequency of visits, posedsome difficulties for both researchers andrespondents. A particular challenge waskeeping the sample intact over the entirestudy period, despite relocation of householdsduring the course of a year or seasonalmigration to different urban or rural locations.Efforts were made to ensure that dropoutsremained at a tolerable minimum. Even undersuch transient conditions, the monthlymonitoring of changes and economic activitieswas carried out with a success rate of over94 percent across all sites, with the remainderconsisting of a few dropout households —those which temporarily left the sampleduring a migration period, or those whichpermanently shifted their residence toanother location. Another challenge was thetense security situation that prevented theresearch team from conducting fieldwork inKandahar altogether, and interrupted anddelayed research in Jalalabad when violentdemonstrations hampered the team’smovement for one month.

In sum, the approach chosen proved to beeffective in facilitating an in-depthunderstanding of households, their activities,and living situations. Researchers were able



to build relationships with respondents andin virtually all cases have been welcomed torevisit households to learn about their lives,the changes they encountered, and how theyadapted. As such, the chosen researchmethodology with its focus on a long-termunderstanding provided an extraordinaryopportunity to observe and document urbanlivelihoods from a household perspective,and it has generated data on which to buildan assessment of livelihood strategiesprevalent among the urban poor in

Afghanistan. However, given that the selectedapproach required the team to work in acomparatively small and not randomlyselected household sample in each city, thesurvey is unable to claim that it isrepresentative; the results reported relateonly to the studied households. Still, they doraise issues and challenges facing a broaderselection of urban poor residents in the studysites. An overview of the characteristics ofhouseholds in the three longitudinal studysites is provided in Table 1.

Table 1. Household sample characteristics in the longitudinal study sites

Kabul Herat Jalalabad

Returned Refugees 12 13 13

IDPs 15 13 9

Long-term residents 13 14 18

Female-headed 11 12 3

Headed by person with disability 4 3 8

Have disabled members (not household head) 4 8 2

Nuclear household 28 27 22

Extended household 12 13 18

Pashtun 8 8 32

Tajik 24 21 7

Hazara 8 4 0

Turkmen 0 4 0

Uzbek 0 1 0

Other 0 2 (Arab) 1 (Pashaie)

Total number of households 40* 40 40*

Average household size 6-7 members 7-8 members 7-8 members

Share of sample members below 18-years of age 65% 64% 59%(N=266) (N=287) (N=322)

*In Kabul and Jalalabad, two households split during the study period. They were subsequently included in thesample as separate units, such that in these two sites, the research was continued with 42 households. Accordingly,the average household size in these two sites is calculated on the base of 42 households.



24 Mc Auslan, P. (2005), Urban Land Law in the Islamic Republic of Afghanistan, An Issues and Options Paper,draft 12 August 2005.

Basic services and social infrastructure playa critical role for human asset building andincreased household and individual wellbeing.There is a huge backlog in service deliveryapparent all over urban Afghanistan. Theproblem is principally shaped by widespreadtenure insecurity and what generally hasbeen termed the “urban land crisis” inAfghanistan.24 It is further fuelled by lack ofmunicipal resources and what appears to bea low commitment to serve dwellers in poorurban settlements. The problem is massiveand has major implications, not exclusivelybut most critically so for the urban poor inthe country, a majority of which reside inurban settlements that are informal and lacklegal status. Zor abad is the local term usedfor these informal settlements — literallymeaning “a place taken by force” — wherepeople enclosed public lands and establishedresidence without seeking official permission.The mushrooming of informal settlements in

Afghan cities has been accelerated bydrought, conflict, displacement and refugeereturn after the overthrow of the Talibanregime. It would be cynical to fault the peopleas they are driven by the basic need forshelter and job opportunities in the city. Theresponsibility rests in a malfunctioning urbanplanning system that has failed to providepublic services, particularly the release ofsufficient land for legal occupation. Theminimal to non-existent services extendedto remote and informal settlements forcehouseholds to develop often burdensome andhealth-threatening strategies in order toaccess infrastructure, compromising theirabilities to achieve sustainable livelihoods.Legal tenure is often the first step to servicedelivery, as well as a key asset contributing tourban livelihood security.

The following evidence taken from theexperience of households living in the cities

4. Land, Services and Social Infrastructure



25 The World Bank (2006), Kabul Urban Policy Notes Series n.2, Washington, DC: World Bank.

studied provide insights as to how peoplecope with these conditions of informality andservice backlog, and how they actually accessbasic infrastructure.

4.1 Land, housing and securityof tenure

The problem of informality of land and housingis widespread but takes on different formsin the various study locations. Most pressingis the situation in Kabul, where the WorldBank estimated that 69 percent of its totalurban area is informal and lacks municipalacceptance, thereby accommodating 80percent of the total city population.25 Thisdevelopment has been very effective inpreventing homelessness in the Afghancapital, but lack of legal status and exclusionfrom service provision in these settlementspose serious problems for inhabitants.

In Jalalabad, the problems are quite similar,although no figures on the degree ofinformality are available. Repatriation ofreturnees from abroad and severe droughtconditions in rural areas led to a constantinflux of people to Jalalabad and eventuallyto the occupation of formerly public spacesduring recent years. Many spacious residentialareas were developed only in the last twodecades and bear no resemblance to theoutdated Master Plan still serving as the mainplanning tool for municipal decision makers.This leaves inhabitants insecure and exposedto the threat of eviction. Nonetheless, therehas been some movement in policyinterpretation recently, with the municipalitydeciding to formalise what used to be arefugee colony (but only after initiallyintimidating residents with bulldozers readyto destroy the entire settlement). Thesuccessful regularisation of the settlement,it has to be said, was made possible onlybecause the neighbourhood is located in aso-called “white area”, i.e. outside the spacecovered by the Master Plan, which gave someroom for negotiation with officials. At the

very least, this example shows that landtitling is a viable option. Affected familieswere willing to sacrifice and invest for a legaltitle, proving that a lasting solution can bethe outcome of a decision-making processthat involves community members, who arethe ultimate beneficiaries. (Every householdwas required to pay a substantial sum of62,000 Afghani — the equivalent of US$1,265— which did not include the resources requiredto rebuild housing on newly assigned plots.)

Herat represents an exception in that tenureinsecurity is not so much a problem with mosthouses erected on legal land. The issue ismore on the shortage of affordable housingor urban land on which to build. This isreflected in the comparatively lower rate ofhouse ownership in Herat and higher numberof rental housing (see Figure 2). Tenureinsecurity is most critical to the estimated400 IDP households, which have been livingin two camps not far from the city-centrefor more than 12 years. Families taking shelterin these makeshift houses are under constantthreat of eviction from the military, whichowns the land.

Likewise, the short-term study sites in Pul-i-Khumri and Mazar have high degrees ofoccupied land with informal status. Mazar,for instance, has a large number of low-income groups settling on lands formerlygrabbed and then redistributed bycommanders mostly on the fringes of town— all of which lack basic services. Pul-i-Khumri, in contrast, is characterised by ahuge degree of tenure insecurity concentratedon the hillsides framing the narrow valley onwhich the city spreads out. The municipalityestimates that about 85 percent of the city’sspace is informal and illegally occupied,reflecting the situation in Kabul. Informalsettlements in both sites are acknowledged,for the time being, by municipalities in theabsence of a clear land policy, although theerection of new buildings in informal areasis prohibited and controlled by localneighbourhood representatives.



With the exception of Herat, the enclosureof public lands and subsequent establishmentof informal settlements all over urbanAfghanistan represents an opportunity forpoor urban dwellers and low-income groupsto self-build and gradually improve on theirhousing, establishing a foundation formanaging their livelihoods in the city. Whentalking about housing, however, two thingsmust be considered when it comes to informalsettlements. Depending on the length ofresidence and the imminence of eviction,housing can be very makeshift and often failsto provide adequate and durable physicalshelter. At the same time, there was evidenceof many single room units erected leading tohigh occupancy rates and over-crowded livingsituations: on average, four to five personsshare one room among the study populationin all three longitudinal research sites.

Forty-four percent of the households studiedmanaged to establish their own housing, while33 percent live free of rent on charity basis.Five percent live in tents and 18 percent onrent (see Figure 2). Proof of house ownershipwas claimed only on rare occasions. Fewpeople possess customary deeds, such asdocuments signed by a local neighbourhoodrepresentative, or other papers indirectlyproving ownership, such as electricity bills

issued by the municipality (kitabche barq),or more often municipal sanitation papers(kitabche safahi). In fact, among the studypopulation, only six households in Jalalabadhad acquired an official title for their landin the wake of the recent formalisation oftheir ex-squatter settlement; 10 in Herat.Not a single study household in Kabul ownsa legal title or can claim a right to the landon which their house stands.

Worth highlighting is the high degree offamilies living free of rent in houses belongingto relatives or friends across sites. Often,family-oriented households sacrifice one oftheir rooms to accommodate another family.In other cases, people are allowed to squatin houses belonging to refugees, who haveyet to return, or households migrating forlabour opportunities. These circumstancesshow the high value attached to socialrelations and community networks as a sourcefor securing shelter, or a means to avoidrental payments that are a heavy burden onalready tight household budgets. Across allstudy sites, it was observed that rents canbe paid only very irregularly by poor urbanhouseholds living in rental housing, leadingto a high dependence on a landlord’sbenevolence and often to unproductiveindebtedness. Rental payments are a constant

Figure 2. Access to housing in the study population (120 households)



19

14

20

9

14

910

12

8

20

3

Have own home Rent housing Live free on charity Live in tents


26 Schütte, S. (2006), Gaining Some Ground: Urban Livelihoods in Jalalabad, Kabul: AREU.27 The World Bank (2006), Kabul Urban Policy Notes Series n.2.

pressure on households relying on irregularand poorly paid jobs, and the inability to payrent on time often leads to eviction andsubsequently to the loss of establishedneighbourhood networks.

Housing certainly represents the mostimportant physical asset a household canpossess in urban Afghanistan — both in termsof saving rent and the potential for using itproductively, but also in providing a footholdin the urban economy. Given a chance, peopleare willing to go into much hardship and riskif only to own their homes. This is apparentin Jalalabad’s recently formalised settlement(see Box 1).26 People will also allocate scarceresources to invest in improving their homes.In Kabul alone, investments in informalhousing are estimated to reach as high asUS$1.3 billion.27

The significance of secure and solid housingto livelihoods adds another critical argumentfor speeding up the search for a solution tothe urgent problem of tenure insecurity.Many households residing on insecure landunderstandably refrain from housingimprovements. Examples where suchinvestments in informal housing had beencompletely lost because of subsequenteviction were encountered in Jalalabad.However, where there is a perceived securityof tenure, as was the case in a newly emergedhillside settlement in Kabul, people put allresources available into the gradualimprovement of their dwellings. In fact,physical asset building in the Kabul case wasa deliberate strategy to secure tenure andto press the municipality for a guarantee tolet people remain on the enclosed land(settlers did not request for a formal title;they asked for the right to stay). It could beargued that tackling the problem of urbantenure in Afghanistan expediently andefficiently may require a general amnestyapplied to informal enclosures. This may beworth considering first, before thinking about

Box 1. Regularising the informal

A sustainable solution to the problem ofinformality in urban settlements and tenureinsecurity was found in one neighbourhood inJalalabad. After residents were able to preventeviction and bulldozing of their houses, themunicipality, in consultation with the Ministryof Urban Development (MoUD) and the localcommunity, decided to implement a land titlingprogramme leading to secure ownership of smallplots of land for urban residents. While this wasonly possible because the settlement is locatedoutside the original Master Plan for Jalalabad,the example shows that finding solutions to theproblem is possible. Households, however, wererequired to pay, but in spite of the 62,000 Afghanineeded to obtain a title and the huge additionalcosts to rebuild houses on allotted plots, theoriginal dwellers succeeded in raising the moneyas a group. This involved hardship, depletion ofphysical assets, taking on high debts and otherrisky money-raising strategies, but people werewilling to go through such adversity to have asecure place to live in. The whole area wasrestructured according to a newly produced mapdelineating new roads, marking the location ofpublic buildings, as well as social infrastructurefor future construction. Individual plots bestowedto each paying household were also identified.Although by the end of the study, people werestill waiting for water access, a health clinic anda school, the construction of boundary wallsdeliniating private space and a slow and gradualreconstruction of housing had commenced. Itneeds to be seen how quickly the municipalityand MoUD will be able to provide basicinfrastructure to the new settlement, but itsprovision had been part of the originalarrangement negotiated. What this positiveexample shows is that a regularisation of theinformal is indeed possible — if official actorsare willing to commit, and if local communitiesare involved in the planning process. Still, thevery precondition for the successful formalisationprocess had been the peripheral location of thesettlement outside the Master Plan of the city.Thinking beyond this central planning instrumentwould allow realising similar projects in otherinformal neighbourhoods. This is a necessary stepto take in order to find a lasting solution to theproblem of informal settlements in Afghanistan— not only in Jalalabad, but also in its otherurban areas.



more complicated, costly and less enforceabletitling procedures. Such approach couldprovide immediate security to informaldwellers and stimulate private investmentsin housing. However, a prior assessment ofthe feasibility of this approach would be inorder, as vested interests in urban land mighthamper such an initiative. This is especiallyobvious when looking at the well-documentedincidences of land grabbing that have takenplace in Kabul and elsewhere. They betraythe shortcomings in governance and the lackof pro-poor policy direction in Afghanistan’scurrent urban politics. What emerges is a“culture of impunity” among those in power,whose self-interest lead them to apply doublestandards to the enclosure of urban land,often at the expense of the vulnerablesettlers. The most prominent example of thisoccurred in the settlement of Sherpur incentral Kabul, where high-ranking governmentofficials received plots of land on which they

built extravagant houses; this after evictingand destroying the houses built by families,who had resided there for decades.28........

This predation on urban land by the powerfuland influential is not an isolated case by anymeans. According to one observer, itrepresents “a microcosm of what has beenhappening all over the city and thecountry.”29 The impoverishment of the bulkof urban populations residing on informalland and the unwillingness to provide themwith services and infrastructure may well gohand in hand with land grabbing that issomewhat legitimised by the fact that theperpetrators are in positions of power. Thisprovides a striking example of the currentlack of governance or mismanagement overurban areas, which begs the question: Howcommitted is the government to promotingequitable development?

28 Constable, P. (2003), in an article written for the Washington Post: “Land grab in Kabul embarrasses government,” 16September.

29 Ibid.



4.2 Basic services: Water supply,sanitation and electricity

Addressing the problem of tenure insecurityin urban Afghanistan must be in conjunctionwith infrastructure development. This isespecially needed in informal settlements,where shortfalls in basic service deliverycreate conditions hazardous to the healthand wellbeing of residents. Although the kindof exclusion from services and infrastructurevary from city to city and even acrosscommunities within a city, the lack of safewater and proper sanitation systems is acommon problem. In Kabul, only 29 percentof the population has access to safe water,with a huge majority relying on shallowgroundwater sources accessed through wellsand hand pumps.30 To make matters worse,groundwater levels are steadily regressingdue to over-exploitation by a risingpopulation. In addition, improper waste andwaste-water disposal result in the tremendous

pollution of such water sources, contributingto the high infant mortality rates in thecountry.31 Situations in Jalalabad (50 percentaccess), and especially Herat (85 percentaccess), are better than in the Afghan capital,although the water supply system in theeastern city of Jalalabad had sustaineddamage over the years through drought, civilwar, and poor maintenance.32

Evidence from the household study revealsthe following patterns and urgent issues:

• Supporting the findings of broader studies,private water access in the studypopulation is a luxury (see Figure 3).Consequently, poor urban households areforced to send members out to haul insupplies, a task often assigned to children.Fetching water is time consuming,especially when the residence is far fromthe nearest public well. If this is not anoption, availing of clean, private waterrequires substantial payments. For hillside

30 GoA (2004), Securing Afghanistan’s Future, p. 54.31 Himmelsbach, T. et al. (2006), Urban Impacts on Groundwater Resources and Quality in the Kabul Basin, Afghanistan,

Geophysical Research Abstracts, Vol. 8.32 Banks, D. (2002), Water Assessment Mission to Afghanistan, January-February, Oslo.



Figure 3. Access to basic services in the longitudinal study sites (40 households in each city)

21

25

20

14

3 4

14

23

6

11 10

5

36

33

29

Have ownelectricity supply

Electricity throughsocial relations

Private water Water throughsocial relations

Have private toilet(dry latrine)


dwellers in Kabul and Pul-i-Khumri, as well as for thenumerous settlers in the desertareas around Mazar, fetchingwater is even more difficult.People in those areas are forcedto buy water from smallentrepreneurs, who turned thewater shortage into a livelihoodopportunity. Some householdscope by collectively sharing adonkey to carry water uphill.In Mazar, people in remotesettlements end up relying ona polluted stream for theirwater supply. In Jalalabad, access to apublic source is closer, but the lines areusually very long, with too manyhouseholds having to rely on a single handpump or public well. Herat, in contrast,has a comparatively high number of poorurban households with access to privatewater, but this usually comes in the formof wells that do not provide potable waterdue to groundwater pollution and soilsalizination. There are a numberof urban poor families in thiswestern city, which rely oninsufficient public sources.About 200 households living inone of the two IDP camps intown share just two publicwells, after six others dried outa while ago.

• In addition to lack of privatewater, access to clean sourcesis highly restricted for manyurban households; this appearsto be especially the case in Pul-i-Khumri and Mazar. The fact that manypoor households in these cities rely solelyon surface water from contaminated riversor streams exposes them to serious healthproblems, particularly young children.There have been incidences of widespreadchronic diarrhoeal diseases amongmembers of poor urban households.

• In many cases, water supply is acquiredwith the help of social relations, whenneighbours with private sources sharetheir well or water tap. This is moreevident in Kabul and Jalalabad, whereprivate water access is more rare than inHerat. Good neighbourly relations haveproven to be essential in accessing water;this is also apparent in the case ofelectricity supply (see Figure 3).

• Proper sanitation systems are absent inthe disadvantaged neighbourhoods acrossall study sites. While a majority ofhouseholds have access to a private toilet(see Figure 3), this is in all cases a drylatrine, where human waste is collectedon the spot without any kind of septicsystem in place. The result is a highly



polluted body of water that flows freelyonto public roads and is channelled intosmall streams of night soil in the middleof streets and alleys. Children play in thishighly unsanitary environment — a perfectbreeding ground for bacteria — makingthem highly vulnerable to chronicdiseases.

• Exclusion from access to electricity iswidespread as well. Many urbansettlements in all cities studied have yetto be connected to local supply networks.But even in neighbourhoods receivingelectricity, poor urban dwellers do notautomatically get access. Running costsare expensive and the set up of wiring,poles and transformer facilities usuallyhas to be financed by communities, asmunicipalities are seldom willing tocontribute. Herat, with the most reliablepower supply among all five study areas,also has the highest number of householdsable to establish private access; notbecause they could afford to buy a meter.

Many of the households studied in thewestern city — where house ownership isless extensive than in other sites (seeFigure 2) — already have electricity intheir residences, given that theiroccupancy is either based on renting orcharity. Neighbours often assist thosewho cannot afford their own electricityby allowing them to tap into their supply.This kind of sharing is more common inKabul (see Figure 3), although somearrangements are reciprocal. For example,a household running a home-based bakeryprovides bread in exchange for electricity.In most cases, however, access is grantedbased on charity.

4.3 Health and education

Delivery of infrastructure includes proximityto adequate basic healthcare and educationalfacilities. This is especially crucial when thereis high reliance on unsafe water sources anda lack of sanitation services that lead tounhygienic living environments and increasehealth risks dramatically. Children are themost susceptible to such conditions. Notsurprisingly, there is a high incidence of youngchildren being regularly sick among thestudied households. There were study familiesin all cities, which lost their children due tounaffordable healthcare and medicaltreatment.

It is in the areas of health and education andthe build-up of human assets that the notionof access needs to be further clarified. Poorurban communities may have a school nearbybut what good is it if children are deployedto work instead of being sent to school? Theremay be a clinic in the neighbourhood, but whatgood is it if families could not afford emergencycare or even the most basic check-up?.....

Across study households and cities, therewere variations in: proximity to adequatelyresourced and affordable social infrastructure,health status and educational attainment,as illustrated in the following sections.



Health

The cities of Afghanistan are the hubs ofhealthcare in the country, with countlessrural dwellers visiting urban centres to accesstreatment on a daily basis. This does notmean that facilities are abundant, or thatclinics or hospitals are well resourced. Eachof the urban sites studied has one or severalgovernment clinics offering free service, butmedicine seems hardly available. For theurban poor, these free but apparently lowquality facilities are the main means ofaccessing healthcare. In addition togovernment services, each city hosts privatehospitals and NGO clinics, both of which areviewed to offer better quality care. However,in spite of all sites providing some form ofhealthcare, facilities remain insufficient. Thecost of consultation in the more reputableprivate clinics is too expensive for the urbanpoor. Moreover, many poor neighbourhoodsare located far away from health facilities,making emergency treatment difficult toobtain.

This combination of distance, low quality andunder-resourced facilities, and unaffordableprivate healthcare prevents the urban poor

in Afghanistan from seeking preventive oremergency treatment.

Susceptibility to poor health, arising out ofexclusion from basic services and the inabilityto access quality treatment, is well reflectedin the survey data collected over the studyperiod. The frequency of spending for healthissues indicates a general predominance ofhealth problems among the surveyed families.A significant percentage of households copewith health-related expenditures almost everymonth. Figure 4 details the frequency ofhealth expenditures in each study site. Thecity of Jalalabad had the highest evidenceof ill health, with medical expendituresreported in 90 percent of the monthlymonitoring interviews across all households;median monthly spending on medicine ortreatment reached 380 Afghanis. In contrast,Herat had the lowest reported incidence ofmedical expenditures, but exhibited thehighest level of spending (450 Afghanis medianmonthly expenditure). The higher cost mayreflect the comparatively better status ofservice delivery in the western city.

The in-depth study in a sub-sample ofhouseholds in each site confirms the



quantitative results. On almost every visit bythe research team, respondents reportedserious health problems encountered bydifferent family members in any given month.Disability, chronic illness and children beingsick on a regular basis lead to huge difficultiesin coping with necessary health expenditures.As a result, urban families fall easily intodebt, or lean heavily on relatives and socialnetworks for support.

It is important to stress that the urban poor’ssusceptibility to health problems is acumulative result of inadequate shelter, useof polluted water sources and the lack ofsanitary waste and sewage disposal systems.Moreover, poor health is a major determinantof household vulnerability. Serious illness ofthe breadwinner, or the health demands ofa sickly child or an aging family member canpush poor households to insurmountable levelsof poverty. This becomes particularly evidentwhen the earner in a one-income family fallsill; the household will have to find ways tocope with the loss of income. If the degreeof income diversification is low, the wellbeingof the entire household can be at risk.

In general, there was no evidence ofpreventive healthcare sought by the householdsamples. The trend was to act only whenconfronted by a health issue, rather than beproactive about healthcare. In many cases,coping with an illness or disease requiresmobilising resources (i.e. money and credit)from social networks. But with the high costof healthcare, poor urban families could notafford to make it a priority.

Figure 4. Frequency of expenditures on health problems (recorded during monthly monitoring interviews)

Box 2. Illness, coping and vulnerability

There are countless examples in all study sites ofchronic sickness and ill health affecting poor urbanhouseholds and seriously hampering their effortsto manage a sustainable livelihood. People do nothave the means to cope with a sudden health shock,more often than not requiring them to go intodebt, to settle an early marriage arrangement fora daughter, or to delay necessary treatment.Chronic illness likewise puts a heavy burden onpoor urban households, with regular medicalexpenditures disrupting tight household budgets.Some examples illustrate the coping strategiespursued when it comes to health issues. The firstexample is from Kabul, where the disabled headof a household, who is a cart puller and singlebreadwinner, was hospitalised after an accident.Suddenly confronted with medical expendituresand total loss of income, the extended familyprovided support. Many visitors to the hospitalbrought gifts and money; his brothers, both mobilevendors also facing insecure situations, providedwhatever help they could; neighbours stepped inand contributed; and the local shopkeeper gavehim a bigger loan to help overcome the difficulttime. Another example is taken from Jalalabad,where the household sold jewellery, which cameas part of the marriage arrangement of its femalehead, to cope with expenditures and income lossafter her husband suffered a blood infection. InHerat, one household settled a marriage for itssix-year old daughter to afford the regularexpenditures required for the treatment of theaging head of the family. A similar strategy wasapplied by a household in Jalalabad, where thebride price for a daughter engaged in her earlyteens is gradually being paid to the household andpresently serves as the means to treat thechronically sick head of household. Some families,however, lack the physical assets to dispose of, ortheir social networks are not able to providesufficient support. This defencelessness have ledto fatal consequences. As revealed during interviewsin Pul-i-Khumri and Mazar-i-Sharif, many familieslost a child to illness, which may have beentreatable, even preventable, if only medicalexpenses were affordable.



Jalalabad 90%

68%

75%

Herat

Kabul

Education

The role of education as a means to livelihoodsecurity is perceived by the study populationas a crucial requirement to overcome povertyin the future. People send their children toschool whenever possible, and the urbancentres of Afghanistan mostly provideeducational facilities. Many study householdsrevealed a preference for practical educationfor their children in the form of skills training,or a combination of apprenticeship and formaleducation. However, there are a number ofproblems hindering many poor householdsfrom providing education for all theirchildren:33

• Some households face a dilemma: sendingchildren to school would mean not havingenough income to make a living in thepresent; on the other hand, deprivingthem of education could mean passingon poverty to the future generation. Thisis the view for 20 percent (N=22) of thehouseholds studied, most of which cannot

afford the outlay of income and timeinvolved in sending their children toschool. Families normally rely on boys foradditional income and on girls to performboth household chores and productivework. In famil ies with femalebreadwinners, daughters are fullyresponsible for household tasks.

• Girls are generally much more likely tobe excluded from education. It is rare forgirls in poor urban households to still beenrolled after they have reached pubertyor after their engagement. This isespecially the case in the city ofJalalabad, where culturally, girls’education is not a priority. Thus, manyhouseholds do not even think aboutsending their daughters to school (seeFigure 5).

• Some 46 percent (N=50) of householdswith school-aged children send at leastsome of them to school (see Figure 6).Elder children are more likely to be

33 Hunte, P. (2006), Looking Beyond the School Walls: Household Decision-Making and School Enrolment in Afghanistan,Kabul:AREU.



excluded because they are relied uponto earn an income or take on householdwork. Younger siblings usually have abetter chance to be educated. There arealso many incidences of childrensuccessfully negotiating between workand school. In fact, a majority of workingchildren also attend school. However, inmany cases attendance is irregularbecause children are made to prioritiseearning an income.

• The proximity of learning facilities is animportant parameter for school enrolmentrates. When the location of schools areat a considerable distance — as is thecase for many remotely locatedneighbourhoods in all cities except forHerat — children are more likely to beexcluded from education. Long distancehas been cited as a major reason whychildren, particularly girls, stay at homein remote communities.

• Thirty-three percent of households (N=36) with school-aged children are able tosend them to school. This is largely dueto a favourable household composition,such as a higher number of male adultworkers. Also, the type of work a

household is pursuing strongly influencesschool enrolment. Those who havemembers with a comparatively securejob and regular income — the exceptionamong the study population — can affordthe opportunity cost for schooling. Thisonly affirms the value of a regular andpredictable flow of income to poor urbanfamilies. For the few households that aredeeply embedded in a network of familyand social relations and who can utilisethe resources of co-residential relatives,having these reliable social assets hasbeen decisive in their ability to provideeducation for their children. Likewise,the health status of potential workersstrongly influences decision making onschool enrolment. Households that sendall children for education usually havesons who work at the same time. Thus,school enrolment does not necessarilymean that children are spared fromearning a living.

• Jalalabad has both the lowest enrolmentrates, as well as the lowest number ofhouseholds allowing education for allchildren. This may be attributed to thedistant location of schools, as well as thelow priority given to girls’ education.



Figure 5. School enrolment rates among study households

Boys Girls

57%

56%

Kabul65%

79%

Herat

Jalalabad37%

58%

Kabul has a comparatively high enrolmentrate for both girls and boys, which canbe linked to the relatively well-placedschool facilities all across the capital city.In Herat, enrolment is relatively weak,

although schools are in closer proximityto most city neighbourhoods. This is dueto the exceptionally high number of bothboys and girls working for an income (seeFigure 7).

Figure 6. Number of households by school-going status of children

Figure 7. Share of school-aged boys and girls working for an income



48%

10%

56%

35%44%

2%


Boys

Girls

All children in school No children in school Some children in school


20

12

45

89 10

18

22

34 Davis, M. (2004), “Planet of Slums, Urban Involution and the Informal Proletariat”, New Left Review, p. 5-34.

5. Urban Labour Markets and Household Consumption

Figure 8. Economic activities of working individuals across the study year(share of reported incidences of work falling under each employment category)



6% 5%

40%

6%

19%

32%

38%33%

1% 3% 3%2%

36%

19%

57%

Regularemployment

Self-employment

Casual wagelabour

Home-basedwork

Others (begging,unpaid work as

helper)


5.1 Urban labour markets andincome-earning opportunities

Informality is the basic feature of urbanlivelihoods in Afghanistan. This is not only interms of land and housing, but also in relationto work and income opportunities. Informalemployment is a lifeline for the great majorityof the urban population. The problem is, asthe main source of livelihoods, the informaleconomy is characterised by irregularity, lowwages, unreliability, high competition andseasonality. In spite of being the single mostimportant source of work, its size is notsufficient to accommodate the steadilygrowing urban population. It has beenelaborated upon elsewhere that globally:

... most informal workers (are) theactive unemployed, who have nochoice but to subsist by some meansor starve.34

This need to create one’s own marginaleconomic opportunities in the informal

economy was clearly articulated by the studysample in the three longitudinal sites. Manylivelihoods totally depend on women andchildren working for extremely low income,with some resorting to begging, experimentingwith new income activities, changing areasof work frequently — all to grasp at everysmall opportunity to earn. Reducing urbanpoverty and vulnerability involves seriouslyimplementing what is advocated in the InterimAfghanistan National Development Strategy(I-ANDS): “unemployment and under-employment are chronic problems that requirea pro-poor growth strategy.”

To be able to afford living in the city, whatkind of employment strategies do poor urbanhouseholds engage in? To answer this question,it would be helpful to examine broaderoccupational categories that fit the studypopulation’s income-earning activities (seeFigure 8), before breaking the categoriesdown into more precise household livelihoodsand discussing issues around the quality ofavailable opportunities (see Table 2).

36 Scoones, I. (1998), Sustainable Rural Livelihoods: A Framework for Analysis, IDS Working Paper 72. Sussex: IDS.



to access a sufficient number of workdaysover the year. Work opportunities are highlyaffected by weather conditions, fluctuationsin demand and the health status of workers,and most respondents faced huge difficultiesin finding work over extended periods. Thisis reflected in the overall low number ofproductive workdays among study householdsacross sites (see Figure 9). In Kabul, 72percent of all workers found employmentonly for less than 200 days a year; in Herat69 percent; and in Jalalabad 74 percent.These values are estimated to be at theminimum levels for creating secure livelihoodsin rural areas.36 Median days of work areconsiderably below this floor. Given thatliving costs in cities are usually higher thanin rural areas, this minimum seems too lowand this incapacity of individuals to find orcreate more working days illustrates thepresent critical state of urban labour marketsin Afghanistan. This is more obvious whenlooking at actual incomes earned vis-à-visnecessary expenditures required to affordbasic household consumption (see Section5.3 and Table 6).

A way to deal with the problems ofseasonality, insecurity of labour and erraticincomes is to diversify income sources bymobilising other household members asearners. Diversification of livelihoods is usuallycarried out through children and in case ofKabul and especially Herat, also women (seeFigure 10). Thus, household structure andcomposition plays a critical role when itcomes to diversifying sources of livelihoodsand the management of urban livelihoods ingeneral (see Section 5.2). Figure 11 depictsthe labour force composition for one year inone study household from Herat. The eight-person household employed differentmembers at different times and frequentlychanged the number of mobilised workers,in response to needs and opportunities, whilestill managing to keep at least three peopleworking regularly. It was largely dependenton female and child labour, with onlyoccasional contributions from the male headof the family. Although the householdmaintained a more stable labour profile, thiswas achieved at the cost of the children’seducation, compromising the next

Figure 9. Median days of work and share of workers employed for less than 200 days per year

Share of Female No maleworkers Overall head of HH labour Pashtun Tajik Hazara Turkmen

Kabul .27 .36 .21 .23 .24 .41 NA

Herat .38 .49 .24 .22 .41 .33 .50

Jalalabad .25 .25 .13 .22 .37 NA NA

Table 3. Mean share of workers in the household



Median days ofwork per year

Share of workersemployed for lessthan 200 days per year

Herat

96

69%

Jalalabad

120

74%

Kabul

131

72%

generation’s opportunities for sustainablelivelihoods.

Differences in shares of workers in thehouseholds across the study sites and otherkey variables of interest further support therelevance of demographic factors andhousehold composition in shaping urbanlivelihood strategies (see Table 3). Householdsin the Herat sample mobilise more labourthan in the other two cities and achieve thehighest median per capita monthly income(see Table 4). When comparing the labourmobilisation strategies of female-headedhouseholds and those without male labour,one can see that female-headed householdsmobilise far more workers than householdson average and than those without maleworkers in both Kabul and Herat. Jalalabadis different, likely due to the strongerinfluence of Pashtun culture (80 percent ofrespondents are Pashtun). This explanation

is supported by the very low labourmobilisation among Jalalabadi householdswithout male labour, illustrating the ratherextreme barriers to women working, evenwhen there are no others in the familyavailable to work. The difference in share ofworkers in the household by ethnicity furthersupports the influence of Pashtun culture onlabour mobilisation, and particularly onmobilising women’s labour.

Figure 10 shows the share of children andwomen in overall work incidences documentedamong the study population in different citiesover the study year, whereby Kabul had thehighest share of working children. Mostoccupations for children, however, involvelow paid petty trades (e.g. shoe polishing,selling chewing gum, biscuits, etc.) orapprenticeships aimed at skills building orvery poorly remunerated assignments as shopemployees. While occupations carried out by

9

8

7

6

5

4

3

2

1

0Feb Mar Apr May June July Aug Sept Oct Nov Dec Jan

Figure 11. Labour mobilisation pattern of one study household in Herat



Figure 10. Share of women and children among all workers

Male adult workers

Male child workersFemale child workers

Female adult workers

Non-workers

0% 10% 20% 30% 40% 50%

Incidences of children working Incidences of women working

Kabul42%

27%

Herat28%

43%

Jalalabad22%

15%

children tended to be similar in all cities,Herat and Jalalabad have altogether lowershares of employed children among allworkers. For Herat, the reason is most likelythe high involvement of women in productiveactivities and a somewhat better economicsituation that leads to a wider array ofinformal job opportunities for the urban poor.

In Jalalabad informal work activities werecharacterised by higher stability; people didnot swap occupations as frequently as theydid in Kabul. There was also high incidence

of seasonal city-to-city migration oversummer, which is when many residents ofJalalabad seek employment in the milderclimate of Kabul. The low share of bothchildren and women contributing to householdincomes, however, led to a very low degreeof livelihood diversification in Jalalabad. Infact, 55 percent of Jalalabadi householdsrely on a single income source and are, thus,highly vulnerable to complete loss of income.The figures for Kabul (27 percent) and Herat(15 percent) indicate less vulnerability — inmost cases, women and children were



Table 4. Median monthly per capita income levels of households with and without a female headand access to male labour

Median monthly per No access to Access tocapita income in: male labour male labour Female head Male head

Kabul 111 Afghanis 494 Afghanis 435 Afghanis 403 Afghanis

Herat 327 Afghanis 678 Afghanis 749 Afghanis 610 Afghanis

Jalalabad 294 Afghanis 580 Afghanis 438 Afghanis 551 Afghanis

responsible for bringing in additional income.Surprisingly, the lower share of childrenworking in Jalalabad does not mean thatschool enrolment rates are higher (see Figure5). In fact, Kabul with the highest incidenceof children working also has the highestenrolment rates.

The degree to which women are involved inproductive labour activities variesextraordinarily between study sites, showingthe clearest difference when comparing theresults in Herat and Jalalabad (see Figure10). In Herat the share of women workingreached surprising heights, making up 43percent of all incidences of work in the localstudy population. This is likely due to labourmarkets being more vibrant in the westerncity, providing more opportunities for womento engage in home-based income activities(see Figure 8). As it is, women’s contributionto income is not only needed, but also viableeven in a conservative social environmentsuch as Herat, where purdah and honour limitwomen’s activity spaces, but do notcompletely constrain their ability to earn anincome.

In Jalalabad women account for only 15percent of all workers and home-based workmakes up a mere one percent of all workactivities. Contrary to Herat, the orthodoxsocial environment hinders women’sengagement in productive activities. Inaddition, typically female dominated home-based activities, such as processing of dryfruits, washing clothes, weaving carpets orspinning wool, do not appear to have a market

in Jalalabad. The few incidences of womenworking in the eastern city virtually all relateto households not consisting of able-bodiedmale members. In Kabul, women wereresponsible for more than a quarter of allrecorded work activities, often also due tounfavourable household compositionsrequiring them to take on paid work, or toaugment the male earner’s income when hisjob is characterised by uncertainty.

Altogether, the findings on female labourparticipation show that women in urbanAfghanistan contribute significantly tohousehold income generating strategies, andit is clear that in spite of prevalent orthodoxviews about women, their access to work isa critical asset and important means ofmanaging urban livelihoods. Yet women arestill generally excluded from urban labourmarkets. Most female dominated employmentis extremely poorly paid that women’s incomehardly justifies the effort required to carryout the job. However, the widespreadevidence of very precarious householdsituations makes involvement of women andchildren in labour activities mandatory.

While labour markets across sites tend tooffer only unprotected, low wage andaltogether insufficient work opportunities,there are significant differences betweencities as to the extent of vulnerability. Lookingat the quantitative data on actual incomegenerating activities documented in the threelongitudinal study sites, Herat by far offersthe best chances to access work. The 40 studyhouseholds in the city engaged in 1,274

different work activities over oneyear. The figures for the other studysites were much lower, in Jalalabad,778, and Kabul, 625. Most of thesedifferences between Herat and theother sites can be accounted for bythe high share of women working,but they also point to thecomparatively higher economicvibrancy of the western city as acentral hub for cross-border trafficwith Iran.

Kabul as the capital city seems milesaway from Herat’s economicposition. This despite the fact that even inthe western city, more informal workopportunities for low-income groups do nottranslate into secure livelihoods. It must benoted that the majority of study householdsin Herat were in highly precarious positionsduring the study.

Nowhere do children compose such a largeshare of the labour force, and nowhere doseasonal fluctuations have a greater impacton income-earning opportunities than inKabul. Even in Jalalabad, where the actualnumber of work activities among studyhouseholds is not much higher and wherelivelihood diversification is very low becauseless women and children join the workforce,the ability to steadily access work isapparently easier.37 The bustling bazaars inthe city are still able to provide low but atleast steady incomes for healthy adult maleworkers in informal occupations. Harassmentof mobile vendors or cart pullers is also muchless in Jalalabad than it is in Kabul. This,together with the smaller impact ofseasonality, likely serves as an importantreason why realised incomes in Jalalabadshow less variance than in the capital city.

5.2 Household structure and thedivision of labour

Household structure and composition arecritical contextual factors determiningvulnerability and livelihood security. Theyare decisive for a household’s ability tomobilise labour and therefore they influencethe possible degree of income diversification.Further, household relations, both internaland external, represent a critical livelihoodasset that governs access to and control overresources for the household and its individualmembers, and this is key to managing thedivision of labour inside the household.Depending on specific household structures— i.e. nuclear or extended, female-headed,those led by persons with disability, thosewithout male labour, or those which formpart of a network of co-residential relatives— the task of managing labour and otherhousehold concerns takes on different forms.

The changing shape of urban households

Urban households in Afghanistan are fluidand flexible, constantly adapting to lifestages, their needs and the environment.

37 It has to be taken into account, however, that due to security restrictions only 11 months of study were accomplished inJalalabad, Therefore, it is possible that the number of work activities among the studied households may be higher thanwhat was recorded.





Such changes in the course of the domesticlife cycle include the marriage of sons anddaughters and the consequent formation ofseparate units, also in response to crises orinternal conflict. In all three longitudinalstudy sites, there was evidence of householdssplitting or merging for various reasons,showing that the adjustment of householdsize, structure and composition serves as astrategy for poor urban households to copewith shock and crisis.

More specifically, the following incidenceshave been documented among the householdsin the study sample.

Households split:

• When a planned transformation from anuclear to an extended unit after themarriage of a son did not go smoothlyand conflict among women aroundhousehold tasks occurred.

• When non-altruism led married sons, alongwith their spouses, to split in order toavoid obligatory contributions to the costsof major health treatments or life cycleevents of family members. This showsthat in the face of low and erratic incomesand poverty conditions in general, peoplemay frequently aim to avoid collectiveburdens that potentially hinder the pursuitof their own opportunities, even if itcomes at the price of putting stress onthe family and jeopardising future sociallinks.

• Out of economic rationality, given thatbigger and extended households are moredifficult to maintain in urban areas. Thisis why in general, smaller household unitsare predominant in Afghan cities, a factthat is also reflected in the shape of thestudy sample (see Table 1).

Households merged:

• As a response to health shocks, when amain breadwinner fell ill or died, aformerly independent household of a sonor daughter rejoined the conjugal unit.

• When a widowed household joined abrother-in-law’s family.

• To deal with economic decline, whenhouseholds failed to generate sufficientincome and needed to sell property(house); they moved in with relatives.

Households adjusted their size andcomposition:

• Out of food insecurity, children were sentaway to relatives in rural areas todecrease dependency rates and to releasepressure on income generation in thecity.

• To enable education for children, whowere sent to school in native villages tostay among relatives.

• When members left the household fortemporary labour migration to aneighbouring country, another city orrural province.

• When rural relatives were accommodatedduring their labour migration period inthe city.

• When sons and daughters of relativeswere accommodated to support incomegeneration and other household activities.

Household structure and vulnerability

It is clear that family structures influencethe capacity of households to manage asufficient income in the city and thus are acrucial determinant for urban livelihoodsecurity. However, the labelling of householdsaccording to different types or groups couldbe misleading. Although female-headedhouseholds often find themselves at lessadvantageous economic and social positions,the conclusion that all female-headedhouseholds form a homogeneous “vulnerablegroup” is not accurate, as the results in Table4 show. In fact, households who are headedby women are not necessarily worse off interms of realised incomes. On the contrary,in the Kabul and Herat study sample, male

Table 5. Median monthly per capita income levels of households with and without members havinga disability

Median monthly Do not have members withincomes in: Have members with a disability a disability

Kabul 260 Afghanis 446 Afghanis

Herat 481 Afghanis 678 Afghanis

Jalalabad 500 Afghanis 570 Afghanis



and female-headed households have similarper capita monthly income, or female-headedhouseholds do better, respectively. It is onlyin Jalalabad where male-headed householdshave the expected advantage.

One important finding of this research is thatthe gender composition of the householdlabour force is a major determinant ofhousehold wellbeing in terms of earnedincome, rather than the gender of the headof the household alone. This is supported byincome data collected over the study year,showing that households having no access tomale labour generate significantly lowerreturns from work activities than those whohave (see Table 4). This is because femaleoccupations are often less valued and do notgenerate fair incomes, and access to workusually is highly restricted for women, makingaffected households particularly vulnerableto income shocks. Herat again is theexception, although there appeared to besufficient opportunities for women to accesspaid work. Only four among the local studyhouseholds did not have women working.

The inclusion of women in the labour marketis often characterised by inequity anddependent relationships with employers. Thisis apparent in the data on median dailyearnings for females and males in the threelongitudinal study cities, where in Heratwomen earn an extremely low daily wage(40 Afs compared to 60 Afs in Kabul and 90Afs in Jalalabad). Thus, even though manywomen hold jobs, their income contributionis insignificant in terms of lifting their familiesout of poverty or increasing their own statusin the household.

Another broad category of households involvesthose who have to take care of disabledmembers or those headed by persons withdisability. Job opportunities are much fewerin these cases as disabled persons usuallycannot compete for wage labour in theinformal labour market and their restrictedmobility makes it difficult for them to engagein many self-employment activities. Theability to mobilise labour of able-bodied sonsis again a critical asset for these types ofhousehold, as is the capacity to activatesupport from social networks. There are casesin the study sample where households headedby persons with disability are highlydependent on intricate social networks orare supported by co-residential relativesliving in the same compound. These strongsocial assets are often needed to accesslabour, to cope with health problems or toreceive support and assistance. This isespecially apparent in Jalalabad, wherehouseholds headed by disabled persons realisethe highest incomes across the three long-term study sites because of their ability toaccess opportunities through membership insocial networks (see Table 5 and Box 5).

In general, however, households with disabledmembers are income-poorer than others, asshown in Table 5, which also emphasises howcritical health is as a household asset forpoor urban families.

Internal household relations and the divisionof labour

Households are not solely a unit of sharing,cooperation and altruism. Internal relationsare often characterised by negotiation,

38 Dupree, N. (2004), “The Family during Crisis in Afghanistan”, Journal of Comparative Family Studies, 35(2):3-11.



bargaining and conflicting behaviour, whichinfluence intra-household arrangements andthe wellbeing of its members (i.e., pooledvs. non-pooled income, cooperative vs. non-cooperative decision making, etc.). Inaddition, there are several socially definedcategories making up the household, notablygender, but also age, generation and maritalstatus that determine different status, rolesand responsibilities.38 As householdsthemselves are microsystems of resourceallocation this hierarchy may well work tothe disadvantage of some members (forinstance when some children are excludedfrom education), and the actual modes oflabour division and quality of internal relationscan be critical factors determining householdvulnerability.

Internal divisions of labour across study sitesare characterised by traditionally ascribedgender roles, with women assigned tohousehold chores and men largely responsiblefor income generation. The fact that manywomen also work for an income, however,has consequences. Often, it means womenhave to carry the double burden of performinghousehold chores, as well as supplementingthe household income. In households wherewomen work extensively, as is the case mainlyin Herat and Kabul, daughters are burdenedwith the household chores, leading to theirexclusion from education. Many women andyoung girls additionally have to deal withrestrictions in terms of their mobility, withtheir movements and behaviour controlledand monitored by men and senior women inthe household. This hampers women’s accessto services and employment across the studysites. Likewise, participation of women ininternal decision-making processes is generallylow, and the patriarchal nature of familiesputs control over income in the hands ofmen.

While it is difficult to assess a general trendin intra-household relations among the diversestudy population, it is clear that the social

institution “household” by no means can beperceived as a completely cooperative unit.Rather, many study households arecharacterised by:

• Competing interests between differentmembers, i.e. when daughters areexcluded from education because theyreach puberty or are required to work inthe house, or when sons are made towork for an income instead of continuingtheir education;

• Conflicts between spouses about marryingoff daughters in order to obtain brideprice; daughters are never consulted whenit comes to deciding about their futurefate, and child marriage turned out tobe a very common feature among moststudy households;

• The need to adjust to external events,leading women and children to enter paidwork;

• Patriarchal behaviour, non-cooperativedecision-making processes;

• Cash resources largely being controlledby men;

• Permeability of boundaries, in that co-residential members of the extendedfamily or kin often influence decision-making processes;

• Undervalued household responsibilitiesand the perception of women’scontribution to household wellbeing as“natural” rather than necessary; and

• Constraints on the freedom of movementof women.

From a livelihoods perspective, these issueshave implications. Non-cooperation, non-pooling and conflict in any given householdmay put the livelihoods of individual membersat risk. Evidence suggests that economicpoverty and a cycle of unemployment may

Table 6. Median per capita incomes and expenditures in three study sites

Median monthly per Median monthly percapita incomes capita expenditures

Median share of total expenses

Food Medicine Fuel

Kabul 409 Afghanis 578 Afghanis 52% 8% 9%

Herat 640 Afghanis 555 Afghanis 58% 11% 9%

Jalalabad 543 Afghanis 600 Afghanis 68% 11% 5%



work to disrupt intra-household relations andtrigger domestic violence.

5.3 Coping with risk: consump-tion, credit and debt

Cash is indispensable in the urbanenvironment, where commoditisation ofvirtually all aspects of life requires poor urbanhouseholds to generate an income that issufficient to afford basic consumption andfood security. This is a challenge for thestudy population across the board. Cost ofliving in Afghan cities is high, and with lowand erratic incomes, cash is not alwaysavailable to feed the family, to pay rent, orto buy fuel or medicine. In fact, the overallfindings on income and consumption —monitored on a monthly basis among the 40study households in each city — clearly showthat on average and over time, the cost forthe most basic needs usually exceeds income,(see Table 6). Again the city of Herat is anexception. It is the only place where theurban poor on average generate sufficientincome to cover their expenditures. This ismost likely due to the higher degrees oflabour diversif ication and incomeopportunities and the large number of womencontributing to household incomes.

Consumption structures among studyhouseholds show the high share of incomerequired for basic food expenses. High sharesof food expenditures in relation to totalexpenses are typically considered a goodindicator for income-poverty, as shares usuallydecrease with a rising income. However, in

the cold winters of Afghanistan food is notthe only factor that puts pressure on budgets.Expenses on fuel and medicine can also puta dent on income. Food expenses, however,always account for more than half of totalexpenditures (see Table 6) but costs for fuelin Kabul and Herat, and medical expensesacross sites, take major shares of monthlyconsumption for poor urban households. Thosewho have to pay rent on top of these basicnecessities are in the least advantageousposition. Many cannot afford to pay houserent in the first place and accumulate debtswith their landlords. For those who do pay,rental expenses have a high share in totalconsumption: 30 percent in Kabul, 28 percentin Herat and 17 percent in Jalalabad.

The income and consumption data collectedclearly point to a central problem for poorurban households: the inability to establishsavings that they can rely on duringemergencies, or to cope with seasonality,income loss and other shocks. In all cities,the numeric evidence shows that living hand-to-mouth is widespread, severely constrainingthe capacity to cope with a health crisis, forexample. Under these conditions, people areforced to obtain cash through other means,which is where access to credit becomes acritical livelihood strategy.

The significance of credit becomes evidentwhen looking more closely at specifichouseholds and their financial flows over theyear. Figure 12 provides a detailed overviewof the experience of one study householdfrom Kabul, which exemplifies the constant

threat of income insecurity and towhat extent households cope withloss of income through access tocredit. This household made mostof its earned income by homeproduction of birdcages, which wassupplemented by the female headgoing out on excursions to Kabul’srural vicinity with a larger groupof females to trade cosmetics fora small profit. However, it isapparent from the figure thatseasonality has a crucial effect onthis household — home productioncame to a complete halt in thewinter because the fine wood usedin this craft cannot stand up to moistconditions. Meanwhile, the female member’soutdoor cosmetic sales had to stop as well.In addition, the female head suffered froma chronic health issue, severely hindering herincome activity. Subsequently, there weretimes in the study period where this householdhad no earned income at all and needed toborrow in order to meet expenditures forbasic consumption. Heavy cuts in consumptionwere implemented as a coping strategy duringvarious months of the study and subsequently,patterns of spending became very erraticwhile the household tried to adjust to its

changing income and flows of credit.

The experience of this household is notunique, and finding access to informal sourcesof credit probably is the most widelyimplemented livelihood strategy in urbanAfghanistan. Virtually all study householdsregularly approach relatives, friends orshopkeepers for a loan, and a great majorityof study households are almost constantly indebt because their income is insufficient topay for their basic needs. There are howeversome differences across study cities, andalthough most households in all sites have to

Figure 12. Financial flows in a study household from Kabul



100%

90%

80%

70%

60%

50%

40%

30%

20%

10%

0%Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan

Expenditures Credit Assets sold Total income

Reasons for outstandingstocks of credit

N = 38 households• Consumption 90%• Healthcare 76%• Rent 18%• House maintenance 18%• Productive assets 15%• Ceremonies 16%• Other 7%

N = 35 households• Consumption 67%• Healthcare 36%• Rent 9%• House maintenance 3%• Productive assets 9%• Ceremonies 15%• Other 9%

N = 36 households• Consumption 72%• Healthcare 69%• Productive assets 11%• House maintenance 17%• Ceremonies 22%• Other 8%

Reasons for currentflows of credit

N = 247 responses• Consumption 64%• Healthcare 21%• Fuel 11%• Rent 7%• House maintenance 3%• Productive assets 3%• Ceremonies 5%• Other 8%

N = 119 responses• Consumption 56%• Healthcare 19%• Fuel 6%• Rent 3%• House maintenance 3%• Productive assets 5%• Ceremonies 9%• Other 4%

N = 195 responses• Consumption 68%• Healthcare 31%• Productive assets 2%• House maintenance and

rent 12%• Ceremonies 3%• Other 11%

Median levels ofindebtedness

18,400 AfghanisKabul

Table 7. Reasons for taking credit among the study population

Herat

Jalalabad

5,000 Afghanis

15,000 Afghanis



City

rely on credit, there are differences in degreesand frequency.

Kabul has the most people relying on constantcredit to afford basic consumption — 56percent of respondents answered affirmativelywhen asked if they had taken credit in agiven month over the study period. InJalalabad, this value reached 41 percent,and in Herat only 25 percent. This is congruentwith findings on incomes and expenditures,and Herat with its relative economicprosperity and high degrees of livelihooddiversification consequently has the lowestfrequency of credit use. However, the figuresare not that much different between citieswhen it comes to current outstanding debtamong sample households. Virtually allhouseholds reported being presently indebtedin varying amounts, often for big family eventssuch as weddings or funerals, or forinvestment purposes such as procuringproductive assets used for income generation,or for maintaining housing. An overwhelmingmajority of the households, however, take a

loan just to afford basic food consumptionand medical expenses (see Table 7).

Sources of credit for study households inalmost all occasions have been informal, andbuying food on credit from shopkeepers andindebtedness to a relative, friend or neighbourwere most frequently reported across studysites (See Figure 13).

Other important findings on credit and debtinclude the following:

• Having strong social assets are needed tohave regular access to credit.

• For poor urban households, residence ina neighbourhood with a heterogeneouspopulation eases access to credit fromneighbours. This is the case in Herat,where settlements are usually morediverse and neighbours serve much moreprominently as creditors (Figure 13).

• Endowment with vertical social relationsis decisive in order to access larger credits

Figure 13. Credit sources for the study population in three cities

Box 4. House rent and the downward spiral of indebtedness

Informal credit is a basic livelihood ingredient for the urban poor. While this fact helps to bridge times of emergency,to afford critical investments in health treatment or productive assets, at the same time, it poses the danger of beingcaught in a downward spiral of indebtedness that is difficult to escape. Examples of this are provided by households thatregularly need to take credit in order to find the money for house rent. There are a number of examples where householdsfailed to cope with the constant need to produce an income high enough to account for rent, on top of basic foodconsumption. Inability to pay, more often than not, leads to eviction, forcing many households to regularly face unwantedresidential mobility, along with a huge burden of debts stemming from unpaid rent. The matter appears especiallypressing in Kabul, where inconsistency of rent payments is highest, but also in the other study sites, where householdsgo into debt to pay for their housing, sell physical assets to afford rent, or are evicted because they are not able to payor cope with rising rent. Most affected households aim to resettle in their old neighbourhood, where informal networksof support or work opportunities for women have been already established. Consider the example of one household livingin Kabul: the head is a bazaar worker. His wife set up a home bakery but fell sick and needed treatment. They werealready indebted to members of the extended family to pay their house rent of 2,000 Afghanis, never knowing how toeventually repay the accumulated amount. A new loan was needed for health treatment, and shortly thereafter, thelandlord raised the rent to 3,000 Afghanis, an amount that was not affordable and for which relatives were not able toprovide further credit. The solid neighbourhood relations they managed to create, the regular gatherings of the womenof the neighbourhood, their self-organised literacy courses, the bakery of the female household head, the nearby schoolfor their three daughters — all these things were seriously endangered. Fortunately, the household managed to find acheaper place not far from their old residence, enabling the family to keep up social relations. Still, the downward spiralof debts is likely to continue, as the insecure and highly seasonal income of the main income earner cannot guaranteethat rent will be regularly paid.



80%

70%

60%

50%

40%

30%

20%

10%

0%Relatives Shopkeeper Friends Neighbours Employer Other NGO


necessary for investments in productiveassets or ceremonial events.

• Repayment of credit often involves muchhardship, but is necessary to keep upsocial relations and to re-access creditfrom the same source. Strategies appliedto repay debts often involve child-marriage to obtain bride price, mobilisingadditional household members for incomegeneration, or simply counting on thepatience of a creditor. Inability to repay

often yields consequences. For instance,one household literally escaped from itsdebts by moving from Jalalabad to Kabul,while another household was raided bythe lender and left bereft of any physicalassets. Often, households living in rentalhousing accumulate debts to a landlordbecause they cannot afford paying rent,in many cases leading to eviction and theloss of neighbourhood networks (seeBox 4).

The poor and vulnerable in urban Afghanistancarry out a wide range of livelihood activities,using a variety of assets over which they havesome control. However, given their limitedtotal assets and the uncertainty of thecountry’s political economy, social assets areparticularly important in preventing ormanaging urban vulnerability. This sectionexamines how and when people are able torely on social assets beyond the immediatehousehold. People’s networks are firstexplored in relation to social support andreciprocity at the neighbourhood and familylevels, and then through a review of some ofthe more strategic attempts at asset buildingthrough collective action and representationwithin the context of local politics and urbanmanagement in Afghanistan.

6.1 Neighbourhood and familynetworks

In the urban context, networks of reciprocityare generally considered to be more fragileand unpredictable than in villages. This hasbeen attributed to the heterogeneity andincreasing fragmentation of Afghanistan’surban populations.39

While this is the casefor some urban dwellers, the present studyshows that there is considerable variationboth between and within cities. The continuedsalience of customary social networks andpatterns of reciprocity apparent in manyurban communities in Afghanistan is supportedby the fact that family and neighbourhoodconnections overlap quite significantly. Forinstance, a large majority of study householdshas relatives living in their vicinity, who actas an important source of assistance. Thisappears to be a coping strategy used by poorurban households in Afghanistan, and theresulting safety net represents the mostreliable defence mechanism against hardshipand multiple challenges.

This informally organised social support canprove decisive in maintaining the livelihoodsecurity of vulnerable households. The presentresearch encountered numerous examples ofthis: relatives employing women to care fortheir children, neighbours providing food andcredit during times of extended income loss,or sharing information about workopportunities. Thus, it is not surprising thatpoor and vulnerable people invest seeminglydisproportionate amounts of time and energyin developing and protecting their socialassets: visiting distant relatives, attendingfestivals together, investing time andresources in supporting people through lifecycle events, such as weddings and funeralsin the confident knowledge that their effortswill be reciprocated. Social support can alsotake the form of community level collectiveassistance, exercised on behalf of membersvictimised by social injustice. This wasillustrated by a settlement in Herat (see Box5).

The informal social networks of the urbanpoor, however, very rarely extends beyondtheir own neighbourhood or circle of relatives.There is very little evidence of bridging socialrelations among the urban poor inAfghanistan, which poses a particularproblem. While the majority of them appearto be endowed with strong social assets thatare indispensable to maintain existinglivelihood outcomes, their networks basicallywork on a survivalist level and the resourcesthat can be acquired through them rarelyallow people to rise above poverty. This isalso the reason why the urban poor’s socialnetworks remain highly vulnerable to overuse:without basic material resources in place,mutual assistance is difficult to sustain andcontinued reliance on social assets is likelyto erode existing networks. Informal social

6. Asset Building through Collective Action and Representation

39 See Beall and Esser (2005).



protection systems may thus quickly reachtheir limits, especially among thosehouseholds forced to pursue the mostprecarious and vulnerable livelihoodstrategies. They are more at risk of over-reliance on such social connections and areless able to reciprocate. As such, the issueof informal safety nets is double-edged. Theircritical importance for urban livelihoodsecurity can hardly be overemphasised, yetunder conditions of poverty the maintenanceof social networks can become a laboriousand sometimes even impossible task. Theavoidance of every collective burden thatpotentially hinders the pursuit of their ownopportunities turns into a deliberate strategyfor many among the urban poor, even if it

40 Landolt, P. (2004), Eine Abwägung der Grenzen Sozialen Kapitals: Lehren aus den transnationalen Gemeinde-Initiativen ElSalvadors, F. Kessl and H.U. Otto (eds) Soziale Arbeit und Soziales Kapital, Zur Kritik lokaler GemeinschaftlichkeitOpladen;Beall, J. (2001), Valuing Social Resources or Capitalising on Them? Limits to Pro-Poor Urban Governance in Nine Cities ofthe South, International Planning Studies.



Box 5. Access to opportunities through socialnetworks and neighbourhood support

The importance of informal support networks isespecially apparent in this example of a householdin Herat, headed by a man caring for his mentallyill wife: they lived a destitute life in a destroyedhouse with a single room without windows,bathroom, or kitchen. Elders of the neighbourhoodmet to address the issue and decided to providethis household with a small room attached to themosque to use as a shop, plus a loan to invest insupplies. It is conveniently located next to a schoolmaking children the main customers. The storeprovided the household with a feasible livelihoodopportunity. When the owner of their residencedecided to reconstruct the building where theysquatted, the family was evicted and the entirehousehold needed to move into the tiny shop fortemporary shelter. Again, the local communitystepped in and allowed them to construct a biggerroom made of mud inside the spacious compoundof the mosque. Another example is provided inJalalabad, where the head of a household is anactive member of the “Afghan Disabled Union”, alocal grassroots organisation. He developed afruitful relationship with a government official inthe course of his voluntary work for fellow disabledpersons. This vertical relation to a powerful personeventually led him to get hold of a piece ofagricultural land on lease, which he sublet. Themoney earned allowed him to open a grocery store,which has become this household’s main source oflivelihood.

comes at the price of putting stress on thefamily and jeopardising future social links.This fragility built into the architecture ofinformal safety nets under poverty conditionsposes a serious risk for those who have torely on the family as their basic fallbacksystem, especially when there is no state-provided or other formal social protectionsystem in place, as is the case in Afghanistan.As such, strong social assets do notautomatically lead to improved well being,and certainly not when generated out of“networks of survival” — a phenomenon thathas been observed in other contexts as well.40

The apparent high levels of resilience of theurban poor thus come at a price and do notrelieve the Afghan government andinternational donor community from theirresponsibility to pursue investments ininstitutionalised social protection systems,employment programmes and physical andsocial infrastructure. The small degree towhich this happened in the past in factcontributed to the present over-reliance ofthe urban poor on their informal socialnetworks.

6.2 Social assets and collectiveaction

The seizure of land and establishment ofinformal settlements is one of the physicalsigns of collective action in pursuit of urbanlivelihoods. When people encroach on publiclands and build without seeking officialpermission, this is rarely done on an individualbasis. Evidence from Kabul and Jalalabadconfirms that land enclosures are the resultof actions undertaken by people known toeach other and loosely organised in someway. Collective action is invariably organisedby groups with something in common, oftenethnicity. In one of the newer settlementsin Pul-i-Khumri — established for 16 yearsand comprising around 110 households at the



time of the research — land was claimedinitially by 10 Tajik families and thereafterencroached upon by other Tajik and Pashaiehouseholds originating from Parwan andKapisa Provinces, respectively. One womannarrated what took place:

Those who came first captured a lotof land. Afterwards, they brought intheir relatives and gave them somespace to build their house on. Someof them also sold land to otherpeople.

Neighbourhood support is not simply confinedto one-off actions such as land grabbing. Inthe process of establishing and maintainingsettlements, there are other institutionalisedsystems of self-help and mutual support asshown in Box 6. In older urban settlements,it was found that residents continued to worktogether to attain services or security oftenure. In Pul-i-Khumri, for example, themost established settlement studied wasbetween 60-80 years old. There, the 1,000or so majority Tajik but ethnically mixedhouseholds had achieved security of tenure;the neighbourhood had even built threemosques, indicating a degree of collectiveaction and representation. Patterns elsewheresuggest that when land titling, upgrading andservice provision occurs, then morecommercialised relationships and transactionstake over, often accompanied by increasingsocial heterogeneity within settlements.

Collective activities can help inject a feelingof commonness and solidarity amongcommunity members, even when theyoriginate from different places and includereturning refugees and IDPs. For example,the study found evidence of people poolingmoney for a school tent or toward the buildingof a mosque. Still, where there was diversitywithin a community, reciprocity was morelimited than in homogeneous settlements.Hence, in a number of contexts it was foundthat self-help systems were delimited bytribal or ethnic boundaries. For instance ina settlement studied in Pul-i-Khumri,mutuality was confined solely to the Ismaili

Box 6. Indigenous systems of self-help: Howashar and pandusi (pandavachi) areused among study households

Indigenous practices of self-help and mutual supportin urban Afghanistan seems to be more widespreadthan the common perception. This is especially soin the case of ashar, which is a reciprocal systemof sharing work in urban areas, most commonly inhouse construction or housing maintenance. Throughthis practice, labour costs are kept at a minimum,since neighbours are required to work withoutpayment – although once given, they have thecustomary right to access labour for their own usein return. While the practice is also used inestablished settlements whenever there is biggermaintenance work required, it is most common innewly emerging settlements where many peopleare collectively busy with the construction ofdwellings. Though this was not exactly the case inthe newly formalised settlement studied inJalalabad, it came close to “starting from scratch”,as everybody living there had to give up their oldhouses and were forced to reconstruct on assignedspaces. As such, there was a lot of activity, andall six households studied from this specificneighbourhood were, at one time or the other,involved in ashar, extending or receiving help fromtheir neighbours to be able to afford rebuildingof housing in the first place. It is fair to state thatwithout this system in place, a number of residentswould not have been able to rebuild so quickly, ifat all, given the high levels of debts and hardshipall of them endure to afford the legal title fortheir land.

Pandusi (also called pandavachi) is very common,although not as widespread. Still, among the 40households studied in Jalalabad, two have takena cow in terms of pandusi from relatives, usingthe dairy products either for their ownconsumption, or for sale. In one case, this is along-term agreement where the taking householdkept the cow for the entire study period.Meanwhile, a new calf was born and the householdextended its physical assets through this system.The second case was not as long: the householdtook a cow from a brother living in the villagebecause he went for a period of labour migrationto Pakistan. Khana Gul, the head of household,had difficulties feeding the cow and spentconsiderable time finding food from natural sourcesin the city or the countryside. This posed highopportunity costs, since he could not be productiveduring those periods. Also, the daily portion ofmilk could not make up for the lost income.Consequently, as soon as his brother came back,he returned the cow, walking all the way with itto the village. The experience was not lucrativeenough to sustain for Khana Gul, showing that, attimes, pandusi may also come as a burden for thereceiver. In this case, however, Khana Gul’s brothergained from the system, as he could be sure thathis livestock had been cared for.

community and none of the Hazara householdsliving there was included in the system. Bythe same token, in Mazar-i-Sharif it was foundthat minority groups, the Hazara for example,were excluded by the bigger groups in thesettlements.

Moreover, collective action is not alwayssustained and it often stands in conflict withthe drive toward individual property rightsthat lie at the heart of urban real estatemarkets, whether formal or informal. Thiscan lead to possible tensions betweeninterventions aimed at land titling on theone hand, and those aimed at communityparticipation in service delivery ormaintenance on the other. Nevertheless, theresearch suggests that where security oftenure is absent, people living in informalsettlements often work in concert to wardoff eviction or lobby to secure land titles.

In spite of all collective efforts, establishingwasita or access to influence in governmentremains a challenge for the urban poor. Inthe absence of social connections to thosein power, people sometimes try to achievesome influence by way of bribery, as hasbeen attempted by a community of returnedrefugees squatting in tents on private landin Kabul. The community pooled a significantamount of money and approached governmentofficials to pursue a process of land allocationin their native area but to no avail. Themoney was taken, but the distribution ofland never occurred. This incident againexemplifies the current degree ofmisgovernance in Afghan cities and thevulnerability of the urban poor to a multitudeof problems that require the attention of areceptive urban government. Attitudes needto change, to counter what was bluntlyexpressed by a higher official in one of themunicipalities of the study sites: “First wehave to look after ourselves, then we canthink about the poor in our city.”

6.3 Social assets andrepresentation

In some contexts relationships based on trustthat extend beyond the household and familyare seen as an indicator of what has beencalled “bridging social capital.”41 This refersto the ability of organised groups or individualhouseholds to link into decision-making andpolitical processes. But as demonstrated inSection 6.1, analysis of the situation inAfghanistan has to be more nuanced, giventhat in many settlement communities, theline between kinship and ethnic affiliationson the one hand, and place of residence onthe other, can be very blurred. Moreover,collective actions on the part of vulnerablehouseholds and communities are oftenreactive and responsive rather than forwardlooking and strategic. When the bonds ofmutuality and reciprocity can be transformedinto civic engagement and representation ofcollective community interests in decision-making arenas, social assets can betransformed into political assets. There is,however, only little evidence of this occurringin the communities studied. In fact, theAfghan Disabled Union from the city ofJalalabad represented the single exampleencountered during the study, where adisadvantaged social group became organised,entered the political arena, linked up withformal governance institutions and articulateddemands. Some of its individual memberswere able to establish vertical social relationstraversing their own social boundaries, whichallowed them to access opportunities forimproving their wellbeing.

However, extending bounded networks ofsocial support to those that can bridge acrossto policy makers, planners and intermediariesis clearly difficult in urban Afghanistan. InKabul, for example, where vulnerablecommunities have suffered displacement,households are subject to insecure tenure

41 Putnam, R.D. (2000), Bowling Alone: The Collapse and Revival of American Community, New York: Simon and Schuster;Woolcock, M. (2001), “The Place of Social Capital in Understanding Social and Economic Outcomes”, Canadian Journal ofPolicy Research, 2(1) Spring, pp. 11-17.



and feel remote from government andpowerless in the face of authority. There islittle evidence of poor households beinginvolved or represented in local level decision-making processes. Moreover, rumours aboundthat when the Master Plan for Kabul isenforced, there would be a clampdown oninformal settlements in the areas covered bythe Plan. Feelings of even greater vulnerabilityand insecurity were expressed by residentsof informal and newly emerg ingsettlements.This development has to be seenin the context of poor levels ofurban governance in the capitalcity, huge backlogs in servicedelivery and a poor track record insocially inclusive urban planningprocesses.

It needs to be acknowledged thatKabul faces particular problems interms of urban governance. As thecapital city and the seat of nationalgovernment, there is evidence ofpoor coordination and evencompetition between KabulMunicipality and the relevant lineministries. NGOs working to deliverbasic services in poor urbansettlements in the city are rarelycoordinated with those agencies planningbulk infrastructure. Moreover, internationaldonors work according to their own deliverypriorities — whether it is the promotion ofland titling, or area-based infrastructuredevelopment projects — which may not bein line with the strategic priorities of theAfghan government. There have been someirregular attempts to coordinate efforts mainlythrough the Urban Management ConsultativeGroup located in the Ministry of UrbanDevelopment. UN-Habitat’s involvement andleadership in this process has been importantin ensuring that lessons on communityrepresentation and management of urbanupgrading programmes have been fed intothis forum, although greater local ownershipof the coordination process is necessary forit to be sustainable over the longer term.

The cultural heritage conservationprogrammes initiated by the Aga Khan Trustfor Culture directly involve local communitiesand have been important both for providingplatforms for local representation and as ameans of demonstrating the value ofcommunity level involvement in decision-making and urban planning.

All urban communities in Afghanistan havean informal council of elders or Shura thatassembles on a regular or irregular basis as

needs dictate. In addition, the wakil-e gozaror neighbourhood representative is recognizedas the official appointee of a community andis deemed to be the one to work closely withthe municipality and speak for communityinterests. Such individuals can potentiallyplay an important role as an intermediarybetween communities and other urban actors,including government officials, privatedevelopers or NGOs. For example, in one ofthe settlements studied in Pul-i-Khumri, aninternational NGO installed a new waterdistribution system in concert with the localwakil-e gozar who, together withrepresentatives of other neighbourhoodsbenefiting from the installation, organisedcommunity participation in carrying outmanual labour, as well as the collection ofhousehold contributions toward the project.



Box 7: Selecting a new wakil-e gozar inPul-i-Khumri

The research team in Pul-i-Khumri had theopportunity to observe first hand a rare occasionin one of the settlements studied — a meeting ofcommunity elders. The purpose of the meetingwas to select a new wakil-e gozar for theneighbourhood to replace the last one after heresigned. The meeting was taken very seriouslybecause the position of wakil is perceived as veryimportant for the community. About 25 peoplefrom the settlement attended, all men. Theyassembled on open public ground and the meetingbegan with a speech by the retiring wakil explainingthe reasons for his resignation. Thereafter,potential successors were put forward and discussedin terms of who would be most appropriate to takeon the role. A number of elders were put forward,alongside a much younger man in his early twentieswho had attained the highest level of educationin the neighbourhood. Many participants felt thathe would be the best choice to represent thecommunity because his education would be anasset. However, he declined the offer saying thathis position would be extremely difficult amongthe elders and that he might not be taken seriouslyby the elderly representatives of otherneighbourhoods and by government officials.Thereafter, there was extensive discussion andgreat effort to convince the second candidate onthe list, a 40 year old man who earned his livingthrough unskilled daily wage labour. Althoughwithout education he was considered honest anddeliberate but he too was reluctant to take on theassignment because of the adverse impact it mighthave on his income-earning activities. It was onlyafter the community expressed their full solidarityand support that he agreed and the following dayhe was introduced to the municipality as the newrepresentative of the settlement. The wholemeeting took a little over half an hour but almostall those present spoke up and the verdict wasreached by way of collective agreement.Nevertheless, women were not present and, norwere their ideas evidently taken into account.

In some cases, members of the Shura workeffectively in advocating for the interests ofthe communities they represent. In othersthey do not. In this case, the system usuallybreaks down. Even when it works well it isoften in a context where women, youngerpeople and other marginalised groups areexcluded. In Pul-i-Khumri, the city does nothave a formal Shura system that could beused to represent community interests, orarticulate community problems to municipalor other government officials. The study also

shows the limitations of the informal selectionprocess for a new wakil-e gozar, as illustratedin Box 7.

Hence the Shura, as a council of elders,constitutes not only an informal system ofsocial support, but also a formal system ofrepresentation by which communities canaccess government decision-making and otheragencies working at the local level. Thereare informal systems to ensure that the wakil-e gozar and the elders are effective and fairbut there are no systematic mechanisms toensure accountability and women’sparticipation. Even when communityrepresentatives are sincere in their effortsto improve the living conditions of theirsettlements through active citizenship, theydo not always get a responsive government,as shown in Box 8, drawn from the researchundertaken in Herat.

Mazar-i-Sharif is fairly unique among Afghancities in that it has functioning electedcommunity councils. They represent theresidents of all 10 city districts and thisarrangement has been in place for more thana decade. The Community ForumDevelopment Organisation (CFDO), a localNGO established by UN-Habitat in the mid-1990s, was responsibile for steering theprocess. Whereas similar community levelorganisations in Kabul and Herat involvethemselves in community-based upgradingactivities, in Mazar, elected Shuras haveconcentrated on addressing the pressing issueof insecurity of income. Hence, most of the199 neighbourhoods that make up the cityhave introduced a variety of skills building,income-generating activities, such asembroidery, tailoring, pottery and baking,as well as projects tackling literacy, computertraining and English classes. In principle,these community councils are inclusive ofwomen and younger people, and in fact manyskills building programmes are headed bywomen and the youth.

The i n t e r v i ewed ne i g hbou rhoodrepresentatives (referred to as Kalantar),which often work in cooperation with the



Shuras in Mazar, also reported that theyrepeatedly went to see government officialsto address issues critical to their communities,such as inadequate water supply, deficient

electricity networks or the absence ofeducational and health care facilities.However, due to lack of capacity andresources the municipality of Mazar is not ina position currently to pursue even minorinfrastructure projects, so that efforts atrepresentation are fairly futile.

Thus, while social assets often lead directlyto community level collective action, theytranslate less easily into representation. Inthe first place, social asymmetries at thecommunity level mean that not all membersare similarly represented, with the interestsof more powerful groups usually prevailing.Moreover, representation involves a two-waystreet, implying a responsive government,which is not always forthcoming in the urbancontext, whether due to resource constraintsor lack of political will. There is overlap andcompetition between national and localgovernment agencies operating in cities andthis leads to a confusing institutional dynamic,where, in a highly politicised environment,the responsibility for urban developmentrests in everybody and nobody. Thus, inexploring the impact of collective action,representation and local governance on urbanlivelihoods, it is imperative that the powerdynamics at work are recognized and takeninto account.

Box 8: Communities and the failure of urbangovernance

Attempts by community representatives to establishlinks with municipal or government authorities inorder to improve living conditions in theirsettlements are not a common occurence. Whenthey do happen, they could turn out to be a futileendeavor. Two examples occurred during a year ofstudying vulnerable communities in Herat. Thefirst relates to a settlement of internally displacedpeople living in the city. They had repeatedly sentdelegations to government authorities in pursuitof security of tenure, but all efforts taken todemand their right to the city were not yieldingresults, and insecurity of tenure remains a seriousand threatening problem for dwellers in the“camps”. The second example refers to acommunity delegation representing a settlementpredominantly inhabited by Hazaras, who contactedthe municipality and demanded the improvementof a short stretch of road that connects theirneighbourhood to the rest of the city. The badcondition of the road posed a serious problem tothe community in which a majority makes a livingby pulling cart or mobile vending, as people cannottraverse easily with their vehicles. However,although the delegation was received and theirproblem acknowledged, no subsequent municipalaction had been taken.



This report reinforces the understanding thatAfghan people are resourceful. They havesurvived decades of war and displacement.In the absence of significant economic activityand responsive governance, they have provento be capable of providing for themselves,both through individual household effort, aswell as collectively. When opportunities forinteraction have presented themselves, theyhave invariably engaged creatively withgovernment. Under such circumstances it canbe tempting for policy makers to continueto rely on the energy and resourcefulness ofvulnerable people, setting aside their ownresponsibilities to deliver even the most basicservices to poor urban residents. One of thekey issues highlighted by this report is thatthere are limits to the resources of the urbanpoor. The disruption characterising the last30 years has significantly depleted theirmaterial assets. The weakness of the nationaleconomy renders them even more vulnerable,with very limited access to employment andincome-earning opportunities and highdependence on borrowing. Moreover, theonerous burden placed on customary patternsof communal support and reciprocityintensifies the pressure on their social assetsand threatens their ability to rely on socialrelations as a safety net in times of crisis.Against these insights, the most importantlesson for policy makers and urban plannersis: while it is important to recognize theresilience and resourcefulness of the urbanpoor, this is not a strategy. Poor peoplealready face many barriers within adysfunctional institutional environment. It isthe responsibility of the state to rectify itsown misgovernance and create anenvironment supportive of the livelihoods ofthe urban poor. This final chapter examinesthe implications of the study’s findings fornational level policy, as well as urban policyand city planning and practice.

7.1 Key findingsThe key findings of the research directlyinform policy implications as enumerated

below:

• Poor and vulnerable urban householdsare exposed to constant variation andchange. This is manifested in highinsecurity of labour and income, whichforces people to constantly adjust andreadjust their occupational activity. Thismeans they have to rely on erratic andshort-term income flows.

• The most vulnerable households alsoregularly shift residence, both in pursuitof work and to avoid steadily rising rents,leading to insecurity of shelter. This is ina context where housing represents themost important physical asset poor urbanhouseholds can aspire to and acquire.

• Household structure and composition isa key determinant of wellbeing — withthose households able to mobilise malelabour being better off than those unableto do so. Urban vulnerability is criticallylinked to health, particularly those ofmale breadwinners. This affects foodsecurity, which influences the health ofthe entire family.

• Labour is the most important asset thatcan be mobilised by the urban poor.Coping with shocks and insecurity leadsmany households to activate additionallabour in the form of working women andchildren, invariably at very lowremuneration. This tends to erode thehealth status and educational attainmentof such households. Moreover, someoccupations for women and children, suchas carpet weaving or begging can be athreat to health and safety.

• Social infrastructure and access to basicservices is highly restricted for the urbanpoor due to a combination of unwillingnessto serve the poor, low resources and lackof capacity among responsible agencies.The lack of access to services affects thecapacity of the urban poor to earnadequate income and acquire thenecessary human assets to take advantage

7. Urban Livelihoods and Implications for Policy



of ava i lab le opportunit ies foradvancement.

• The asset base of many vulnerable urbanhouseholds is not sufficiently diversifiedto lift them out of poverty in any lastingway. Their reliance on scant resourcesputs pressure on their coping mechanisms.This applies to households that dependon paid labour, as well as those that relyon the goodwill and support of others.Social networks are vulnerable to overuseand tend to deteriorate, if not collapse,when rel ied upon extens ively.

• There is little evidence of poor andvulnerable households being involved inlocal level decision making processes.This relates to the control of communitylevel organisations by male elders, whodo not always represent the moremarginalised members. Confounding thesituation is the poor quality of urbangovernance currently being exercised inAfghan cities. The latter is evident notonly in the huge backlogs in servicedelivery but also in terms of the absenceof socially inclusive or even consultativeurban planning.

7.2 Implications for nationalpolicy

The most critical policy issue drawn from theresearch focuses on the economy and onstrategies to stimulate broad-based labourintensive growth at the national and citylevels. The acute insecurity of income, thehigh level of dependence on poorly paidcasual work, as well as the proliferation oflivelihoods pursued across the informaleconomy are all evidence of an economy inwhich the potential for urban job creationremains limited. This must change. Possiblepolicy approaches include:

• Evaluating Afghanistan’s more openmarket approach against the possibleeconomic and employment gains fromimport substitution and protectionism in

some key industries, such as cement andcotton/textiles.

• Ensuring that opportunities for goodquality work are valued as much as thequantity of jobs created. Thus, a holisticstrategy supporting urban employmentcreation must include market analysis ofdemand for higher value products andthe labour and skill requirementsassociated with them. This must becombined with the development ofappropriate vocational training and formaleducation programmes to develop a labourforce with the appropriate skills.

• Developing a broad planning frameworkthat anticipates the continuing increasein urban populations, resulting from theinflux of returnees from neighbouringcountries, as well as from pull factorsdrawing rural residents to urban areas.

• Facilitating the participation of womenin the urban labour force under improvedconditions, where their work is valuedappropriately and they are given a realchoice about place of work. In addition,the burden of household work must bereduced by supporting community childcare initiatives and improving access tobasic services.

A second critical national policy issue is toredress the imbalance caused by the presentpredominantly rural focus of developmentassistance in Afghanistan. The implementationof genuine nationwide programmes isimperative in order to combat urban as wellas rural poverty on a scale that will haveimpact. The efforts of the National SolidarityProgramme (NSP) to support livelihoods andto enable poor people to better manage riskare fairly well advanced but its location inthe Ministry of Rural Rehabilitation andDevelopment suggests a rural bias. Urbanpoverty is a complex and growing problemin Afghanistan that requires a policy responsenow. Recommended responses include:.......

• Designing and implementing nationwideurban poverty reduction programmes toimprove the quality of life of the urban



poor at a scale approaching the realityof the problem. These programmes mustrecognisze the inter-linkages betweenthe myriad problems underlying urbanpoverty and take an integrated versussectoral approach, linking job creation,skills building and urban service provision,as well as improving urban governance.

• Streamlining coordination and cooperationbe tween l i ne m in i s t r i e s andmunicipalities. Much of the backlog inservice provision and barriers to pro-poorurban development derive from theinability and unwillingness of governmentagencies to work together. Incentivesmust be developed to ensure effeciency.Successful coordination structures fromother countries could be studied andadapted in developing effective urbangovernance institutions in Afghanistan.

• Increasing donor support for thedevelopment and implementation ofurban poverty reduction and capacitybuilding programmes within governmentinstitutions. Supporting sustainable urbandevelopment requires medium to longterm financial commitments, not short-term funded pilots that are not broughtto scale. Donor support must becomprehensive to achieve successfulinstitutional reforms.

Another national level policy implication isthe need to increase the human capital ofthe urban poor through improved access toquality healthcare and education. In manyof the settlements studied, government clinicswere limited in number or in their ability toadequately serve the growing numbers ofpoor urban residents. Distance to schools andthe quality of education were also key factorsthat bring down enrolment rates of girls andboys. These outcomes lead to the followingpolicy needs:

• Extend affordable and proximate health-care facilities into or close to poor urbanneighbourhoods, allowing low incomegroups to regularly access bothpreventative and curative care.

• Plans for the allocation of new schoolsmust be made in a way that distributesaccess equitably across urban locationsto promote school attendance.

• Efforts to provide proper school buildingsshould be accompanied by greater effortsto ensure quality teaching and teachingsupport.

A final national policy requirement is todevelop risk mitigating social protectionprogrammes, supporting a shift away fromreliance on the coping capacity andresourcefulness of the urban poor themselves.Their asset portfolios are not sufficientlydiversified to protect households againsteconomic trends and shocks or to lift themout of poverty in a lasting way. Thus, it isthe state’s responsibility to protect the mostvulnerable and to promote assetdiversification to reduce livelihood insecurity.The following are recommended to fulfill thisresponsibility:

• Develop and finance national socialprotection programmes to support themost vulnerable (families without ablebodied male workers, families dependenton disabled members for income, etc.)and to invest in programmes supportingasset diversification for the poor to assistthem out of poverty.

• Provide donor support for urban povertyreduction programmes that go beyondphysical upgrading and technical serviceprovision to address social insecurities.

• Build the political will and capacity ofnational and city level actors (bothgovernment and non government) toimprove their abilities to address socialprotection issues at the required scale inthe urban sector.

In sum, when poor urban households do nothave the asset base to withstand stress, thenprecautionary efforts aimed at reducing theimpact of potential shocks rest with nationallevel policy. Similarly, when household copingstrategies do not enable families to recoverfrom shocks, then reactive solutions designed



to soften the impact of adversity also restwith the national government, as well asinternational development agencies involvedin both relief and development.

7.3 Implications for urban policyand city planning

The biggest challenge facing urban policyand city level planning in Afghanistan isnegotiating a path between the reality ofburgeoning informality and the rigidity offormal master planning across all its majorcities. The commitment to Soviet-style urbanplanning is understandable among thosetrained in this way. However, master planninghas been discredited due to its top-downnature, inflexibility and lack of ownershipamong city residents. Instead, more flexibletools that are responsive to the diversity ofurban needs and voices need to be adopted.This is particularly true in urban contextscharacterised by informality — whether ineconomic activity or the development ofsettlements — where flexibility and inclusionof the marginalised are necessary to achievemore equitable urban development. Thus,the first recommendation involves:

• Building capacity among municipal leadersand staff to move them away from asolely technical planning approach linkedto a master plan. Help them accept theimportance of including citizens in theplanning process and to develop moreflexible tools reflecting the on-the-groundreality of all urban residents, particularlythe poor.

Another vital point for city-level action,supported by national policy is the recognitionand legalisation of informal settlements. Itis important that policy makers value thehuge housing investments informal settlementdwellers have made and recognize their “rightto the city”. Continuing threats of evictionand failure to engage in negotiations overupgrading and tenure security are

unacceptable, especially when made in thename of moribund master plans. Requiredactions include:

• Developing a national urban land policyin line with the aim of equitabledevelopment. Establish guidelines andcategories for municipalities that wouldfacilitate recognising and legalisingexisting informal settlements. The policymust also consider how to accommodatethe minority living on untenable land.

• Rectifying past land grabs to ensure thatthe perpetrators are stripped of“ownership” while protecting currentresidents from immediate eviction.

• Providing security of tenure to residentsin recognized informal settlements, withrecommendations for approaches to landtitling coming from an analysis of bestpractices in other similar post conflictand Islamic contexts.

Informality and reliance on master plans thatdo not recognize informal settlements areoften the reasons given for not extendingservices to certain areas. Thus, establishingclear legal status for informal settlementsand including them in future planningprocesses are the first steps on the long roadto ensuring universal coverage with basicservices (water, drainage and sanitation,electricity, access roads). This will not be asimple task due to the range of line ministriesinvolved, the competing roles of municipalitiesand other government institutions in serviceprovision, the resources (human and financial)required, and the difficult issues of servicestandards, cost recovery and subsidies.However, the benefits to the urban poor ofimproved access to basic services will besubstantial in terms of improved health statusand productivity. Moving incrementallytoward universal coverage will require:

• The political will among state institutionsto serve the poor.



• Improved communication betweengovernment institutions involved in servicedelivery, including a detailing of rolesand responsibilities and establishing ameans of coordination.

• Improved communication and coordinationbetween government and non-governmentagencies involved in service provision,for better distribution of responsibilitiesand to share lessons learned and bestpractices.

• Community consultation, involving bothwomen and men, to devise appropriateservice standards and allocate operationand maintenance responsibilities.

• Technical creativity in devising ways todeliver services to some harder to reachareas, ensuring that the work of disparateagencies links up at the city level.

• Recognition of the employment creationpotential of service delivery andinvolvement of poor urban residents(women and men) as small scaleentrepreneurs.

There are important opportunities at citylevel for recognising the importantcontribution of informal economic actors tothe city and not associating “informal” with“backward”. This is particularly the case forvendors, street sellers and cart pullers, whoare among the most excluded informal actors.Some suggestions to rectify this include:

• Vendors and street sellers and their rolein the urban market should beacknowledged, not as a nuisance, but asimportant contributors to the economiclife of the city. Recognition can comethrough some form of licensing,transparently developed to avoid chancesof unscrupulous rent-seeking.

• The needs of street vendors and theimportance of local markets must berecognized in city plans; they must beplanned for and not marginalised.

• Institute an inclusive consultation processto discuss how to regulate informaleconomic activity and to ensure that theneeds of informal actors and themunicipalities are represented. Thismeans ending harassment and rent-seeking and guaranteeing that vendingactivities do not impinge on the effectiveoperation of municipalities.

7.4 Conclusion

The need for cooperation and collaborationbetween different government and non-government agencies has been raisedrepeatedly in this report. Its importance formaking any progress on urban povertyreduction cannot be stressed enough.Governance-related issues relevant to urbanpoverty reduction include:

• Establishing more transparent processesof serving the poor that reduceopportunities for rent-seeking.

• Developing systems to hold all agenciesinvolved in the urban sector accountablefor positive results.

• Involving representatives of poorcommunities in consultations, at the veryleast, and in other more substantial ways( i .e . through employment) asappropriate......

It will be difficult not to have multipleagencies involved in urban poverty reductionprogrammes because of the range of expertiserequired. Government officials and privatecontractors often have the technical expertiserequired for programme delivery, while NGOsand community-based organisations are betterequipped to undertake needs assessmentsand to mobilise urban residents into decisionmaking and cooperative processes. Similarly,there are complex negotiations to be hadwhen pursuing security of tenure throughland titling. This is especially so in the wakeof displacement, multiple occupation of landand repair and extensions to property. The



difficulties involved in consultative andcollaborative planning, whether aroundservice provision and maintenance or landallocation and titling, are often considerable.They require willing and cooperativeinteraction between residents, city officials,private sector developers and local andinternational agencies. However, the researchsuggests that it is sometimes very difficultfor city governments to work effectively andin concert with the Ministry of UrbanDevelopment and other line ministries, suchas those concerned with roads and transport.Their interactions tend to be disrupted bymisunderstandings or competition overresources, or complicated by differentapproaches to common problems. Theinstitutional environment is rendered evenmore complex by the presence of a widearray of international development agenciesintervening in the urban context, rangingfrom the World Bank, UN-Habitat and anumber of bilateral funders, to internationalNGOs such as the Aga Khan Trust for Cultureand CARE. Coordination and cooperationamong them and with national ministries andmunicipal governments are critical for theachievement of effective urban policy andcity planning.

The application of a livelihoods perspectiveto urban settings can serve to obscure thesocial asymmetries at work within individualsettlements and citywide, as well as thecomplexity of the institutional context inwhich both local and national politics andpower dynamics are played out. This is notthe intention of this report. On the contrary,an important aim has been to identify howthe presence or, more often, absence ofurban governance has influenced the assetportfolios of the urban poor. In the case ofAfghanistan, our findings suggest that thelivelihoods of the urban poor have beendeeply affected by the inability of nationalministries and municipal governments tomanage the country’s cities and the rapidsocial, political and economic changes theyface. When they are well managed, city

governments can mitigate the need for ex-ante and ex-post strategies in relation tourban livelihoods, through preventativemeasures that protect the urban poor, forexample in relation to service delivery, publicsafety and the liveability of the environment.If urban policies and city planning are to yieldreal results, they have to take place in acontext where city governments havesufficient powers, resources and staff to beable to deliver services. There needs to bea spirit of collaborative planning andpartnership involving local citizens and theprivate sector, as well as cooperation betweendifferent levels of government. Urbangovernance and city planning that issupportive of the livelihoods of the poorcannot evolve in a climate of mistrustbetween citizens and those who govern them.Similarly, accountability must replace thecurrent practice of finger-pointing and blame-game between those responsible for servicedelivery. Achieving these goals will take timeand requires a great deal of effort andcollaboration among stakeholders. It is hopedthat the energy and resilience of the poorand vulnerable urban dwellers of Afghanistanwill provide the inspiration to alleviate urbanpoverty. They deserve nothing less.



Banks, D. 2002. Urban Water Supply Systems in Afghanistan. Report on Water AssessmentMission to Afghanistan. Kirkens Nødhjelp, Oslo.

Beall, J. 2005. Urban Livelihoods. Towards A New Map of Africa, C. Toumain and B. Wisner(eds). London: Earthscan.

Beall, J. 2001. “Valuing Social Resources or Capitalising on Them? Limits to Pro-Poor UrbanGovernance in Nine Cities of the South”. International Planning Studies.

Beall, J. and D. Esser. 2005. Shaping Urban Futures: Challenges to Governing and ManagingAfghan Cities, AREU Issues Paper Series. Kabul: AREU.

Bebbington, A. 1999. “Capitals and Capabilities: A Framework for Analysing Peasant Viability,Rural Livelihoods and Poverty.” World Development. 27(12):2021-2044.

Carney, D. 1998. “Implementing the Sustainable Livelihoods Approach.” Sustainable RuralLivelihoods, What Contribution Can We Make?, D. Carney (ed). London: Department forInternational Development.

Chambers, R. and G. Conway. 1992. Sustainable Rural Livelihoods: Practical Concepts forthe 21st Century. Institute of Development Studies Discussion Paper No. 296. Brighton: IDS,University of Sussex.

Constable, P. 2003. “Land grab in Kabul embarrasses government.” Washington Post.

Davis, M. 2004. “Planet of Slums, Urban Involution and the Informal Proletariat.” New LeftReview.

Dupree, N. 2004. “The Family During Crisis in Afghanistan.” Journal of Comparative FamilyStudies, 35(2):3-11.

Ellis, F. 2000. Rural Livelihoods and Diversity in Developing Countries, Oxford: OxfordUniversity Press.

Government of Afghanistan (GoA). 2005. Afghanistan Statistical Yearbook 2005. Kabul: GoA.

GoA. 2004. Securing Afghanistan’s Future. Kabul: GoA.

Grace, J. 2003. 100 Households in Kabul: A Study of Winter Vulnerability. Kabul: AREU.

Himmelsbach, T. et al. 2006. “Urban Impacts on Groundwater Resources and Quality in theKabul Basin, Afghanistan.” Geophysical Research Abstracts, Vol. 8.

Holzmann, R. and S. Jørgenson. 2000. “Social Risk Management: A New Conceptual Frameworkfor Social Protection and Beyond.” Social Protection Discussion Paper, No. 0006. WashingtonDC: World Bank.

Hunte, P. 2006. Looking Beyond the School Walls: Household Decision-Making and SchoolEnrolment in Afghanistan. Kabul: AREU.

References



Hunte, P. 2004. Some Notes on the Livelihoods of the Urban Poor in Kabul, Afghanistan.Kabul: AREU.

Landolt, P. 2004. Eine Abwägung Grenzen Sozialen Kapitals: Lehren aus den transnationalenGemeinde-Initiativen El Salvadors. Soziale Arbeit und Soziales Kapital, F. Kessl and H.U.Otto (eds). Zur Kritik lokaler Gemeinschaftlichkeit, Opladen.

Mc Auslan, P. 2005. Urban Land Law in the Islamic Republic of Afghanistan. An Issues andOptions Paper, draft.

Moser, C.O.N. 1998. “The Asset Vulnerability Framework: Reassessing Urban Poverty ReductionStrategies.” World Development 26(1): 1-19.

Moser, C.O.N. 1996. “Confronting Crisis. A Summary of Household Responses to Poverty andVulnerability in Four Poor Urban Communities.” Environmentally Sustainable DevelopmentStudies and Monograph Series, No. 7. World Bank.

Ministry of Rural Rehabilitation and Development. 2003. Nationwide Risk and VulnerabilityAssessment (NRVA). Kabul: MRRD.

Putnam, R.D. 2000. Bowling Alone: The Collapse and Revival of American Community. NewYork: Simon and Schuster.

Rakodi, C. and T. Lloyd-Jones (eds). 2002. Urban Livelihoods. A People-Centred Approachto Reducing Poverty. London: Earthscan.

Satterthwaite, D. 2000. Editor’s Introduction: “Towards more Pro-poor Local Governmentsin Urban Areas.” Environment and Urbanisation, 12(1):3-11.

Schütte, S. 2006. Dwindling Industry, Growing Poverty: Urban Livelihoods in Pul-i-Khumri.Kabul: AREU.

Schütte, S. 2006. Gaining Some Ground: Urban Livelihoods in Jalalabad. Kabul: AREU.

Schütte, S. 2006: Poor, Poorer, Poorest: Urban Livelihoods and Vulnerability in Mazar-i-Sharif. Kabul: AREU.

Schütte, S. 2006. Poverty Amid Prosperity: Urban Livelihoods in Herat. Kabul: AREU.

Schütte, S. 2006. Searching for Security: Urban Livelihoods in Kabul. Kabul: AREU.

Schütte, S. 2004. Urban Vulnerability in Afghanistan: Case Studies from Three Cities. Kabul:AREU.

Scoones, I. 1998. “Sustainable Rural Livelihoods: A Framework for Analysis.” IDS WorkingPaper No. 72. Sussex: IDS.

Sen, A.K. 1999. Development as Freedom. Oxford: Oxford University Press.

United Nations Development Programme. 2004. Afghanistan National Human DevelopmentReport: Security with a Human Face. New York: UNDP.



UNDP. 2003. Human Development Report: Millennium Development Goals: A Compact AmongNations to End Human Poverty. New York: UNDP.

Wood, G. 2003. “Staying Secure, Staying Poor: The Faustian Bargain.” World Development31(3): 455-71.

Woolcock, M. 2001. “The Place of Social Capital in Understanding Social and EconomicOutcomes. Canadian Journal of Policy Research, Spring 2(1): 11-17.

World Bank. 2006. Kabul Urban Policy Notes Series. Washington, DC: World Bank.

World Bank. 2005. Afghanistan: Poverty, Vulnerability and Social Protection, An InitialAssessment. Human Development Unit, South Asia Region, Report No. 29694-AF. Washington,DC: World Bank.

World Bank. 2004. Afghanistan: State Building, Sustaining Growth and Reducing Poverty, ACountry Economic Report. Poverty Reduction and Economic Management Sector, South AsiaRegion, Report No. 29551-AF. Washington, DC: World Bank.

World Bank. 2001. Social Protection Sector Strategy: From Safety Net to Springboard.Washington, DC: World Bank.



June 2006 Going to Market: Trade and Traders in Six Afghan Sectors, by Anna Paterson

June 2006 Management, Livestock and the Opium Economy: Land Tenure,by Alec McEwen and Brendan Whitty

June 2006 Management, Livestock and the Opium Economy: Opium Cultivation in Kunduz andBalkh, by Adam Pain

June 2006 Water Management, Livestock and the Opium Economy: Animal Husbandry,by Anthony Fitzherbert

June 2006 Poverty Amid Prosperity: Urban Livelihoods in Herat, by Stefan Schütte

June 2006 Informal Credit Practices in Rural Afghanistan, Case Study 1: Herat, by Floortje Klijn

May 2006 Water Management, Livestock and the Opium Economy: Baseline Survey, by Alan Roe

May 2006 Water Management, Livestock and the Opium Economy: Social Water Management,by Jonathan L. Lee

May 2006 Water Management, Livestock and the Opium Economy: Irrigation Systems,by Ian McAllister Anderson

May 2006 Continued Protection, Sustainable Reintegration: Afghan Refugees and Migrants in Iran,by M.J. Abbasi-Shavazi and Diana Glazebrook

May 2006 Provincial Governance Structure in Afghanistan: From Confusion to Vision?,by Sarah Lister and Hamish Nixon

May 2006 Gaining Some Ground: Urban Livelihoods in Jalalabad, by Stefan Schütte

April 2006 Searching for Security: Urban Livelihoods in Kabul, by Stefan Schütte

March 2006 Afghan Returnees from NWFP, Pakistan to Nangarhar Province,by Gulbadan Habibi and Pamela Hunte

March 2006 Looking Beyond the School Walls: Household Decision-Making and School Enrolment inAfghanistan, by Pamela Hunte

March 2006 Dwindling Industry, Growing Poverty: Urban Livelihoods in Pul-i-Khumri,by Stefan Schütte

Jan 2006 Poor, Poorer, Poorest: Urban Livelihoods and Vulnerability in Mazar-i-Sharif,by Stefan Schütte

Jan 2006 Afghans in Quetta: Settlements, Livelihoods, Support Networks and Cross-Border Linkages,by the Collective for Social Science Research

Jan 2006 Afghans in Peshawar: Migration, Settlements and Social Networks,by the Collective for Social Science Research

Jan 2006 Afghans in Pakistan: Broadening the Focus, by the Collective for Social Science Research

All AREU publications can be downloaded fromHard copies are available at AREU’s office in Kabul.

Afghanistan Research and Evaluation UnitFlower Street (corner of Street 2)Shahr-i Naw, Kabul, Afghanistan

email: [email protected] website: www.areu.org.af

Recent Publications from AREU



afghanistan research and evaluation unit - · pdf filedeveloping world (2004); ... the...

Documents