The Next AscentAn Evaluation of the Aga Khan

Rural Support Program, Pakistan

2002THE WORLD BANK

Washington, D.C.

W O R L D B A N K O P E R A T I O N S E V A L U A T I O N D E P A R T M E N T

http://www.worldbank.org/oed

© 2002 The International Bank for Reconstruction and Development / The World Bank

1818 H Street, NW

Washington, DC 20433

All rights reserved.

Manufactured in the United States of America

First Printing September 2002

1 2 3 4 03 02

The findings, interpretations, and conclusions expressed here are those of the author(s) and do not necessarily reflect the

views of the Board of Executive Directors of the World Bank or the governments they represent.

The World Bank cannot guarantee the accuracy of the data included in this work. The boundaries, colors, denomina-

tions, and other information shown on any map in this work do not imply on the part of the World Bank any judgment

of the legal status of any territory or the endorsement or acceptance of such boundaries.

Rights and Permissions

The material in this work is copyrighted. No part of this work may be reproduced or transmitted in any form or by any

means, electronic or mechanical, including photocopying, recording, or inclusion in any information storage and retrieval

system, without the prior written permission of the World Bank. The World Bank encourages dissemination of its work

and will normally grant permission promptly.

For permission to photocopy or reprint, please send a request with complete information to the Copyright Clearance

Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750-4470, www.copyright.com.

All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher,

World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org.

Cover photo: Courtesy of the Aga Khan Foundation/Jean-Luc Ray.

Picture taken in 2000 of members of the Sost Women’s Organization gathering fodder on land they were just beginning

to cultivate in 1988. Although it turned out to have soil that was too poor for fruit trees (which they subsequently moved

to better, sloping land), they continue to grow poplar and willow trees for fodder as well as alfalfa on this Women’s

Organization common land.

ISBN 0-8213-4979-1

Library of Congress Cataloging-in-Publication Data has been applied for.

Printed on Recycled Paper

i i i

Contents

v Acknowledgments

vii Foreword, Prefacio, Avant-Propos

xi Executive Summary, Resumen, Résumé Analytique

xxix Abbreviations and Acronyms

1 1. Introduction01 AKRSP Program Objectives02 Methods03 Findings of Previous Evaluations

5 2. Overall Program Assessment05 Relevance07 Efficacy08 Efficiency13 Institutional Impact17 Sustainability

19 3. Community Organizations19 Relevance21 Efficacy23 Efficiency

25 4. Programs25 Microfinance28 Productive Physical Infrastructure30 Natural Resource Management33 Marketing and Enterprise Development

37 5. Future Directions37 The Mission Is Valid, the Direction Needs to Change41 Future Strategy—Community Organizations

Annexes047 Annex A. AKRSP Skills Seeding049 Annex B. Responsiveness to Earlier OED Recommendations051 Annex C. AKRSP Cost Comparators055 Annex D. AKRSP Timeline059 Annex E. Institutional Survey063 Annex F. Villages Visited065 Annex G. Computation of Subsidy Dependence Index for the

Microfinance Operations of the AKRSP067 Annex H. AKRSP Economic Analysis073 Annex I. The Karakoram Agricultural Research Institute for the Northern

Areas (KARINA)075 Annex J. Statistical Tables099 Annex K. Organizational Charts101 Annex L. Main Programs123 Annex M. Recent Donor Program Evaluations of the AKRSP: A Summary

of Main Conclusions

129 Endnotes

137 References

Boxes02 Box 1.1. AKRSP Operations and Their Institutional Setting06 Box 2.1. Unique Political Status of the Northern Areas44 Box 5.1. The Cascading Impact of a New Vision

Tables27 Table 4.1. Efficiency Comparisons32 Table 4.2. Performance in NRM: Strengths and Weaknesses

Figures11 Figure 2.1. Expenditure Categories as a Percentage of Total38 Figure 5.1. Future Share of Development Capacity45 Figure 5.2. Pakistan: The Aga Khan Rural Support Program

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

i v

v

Acknowledgments

This report, and the mission on which itwas based, could not have been com-pleted without the exceptional support

provided by the Aga Khan Foundation and theAga Khan Rural Support Program (AKRSP). Thearrangements for the mission in the field wereexcellent. Moreover, subsequent to the mission,the AKRSP has been unstinting in its provisionof further data and analysis related to elementsof the economic analysis. The very useful inputof the AKRSP Board and the Donor LiaisonGroup is also acknowledged. We would like tothank the teams in the Core Office, in eachregion, and in the Field Management Units, aswell as the helicopter pilots. Thanks are due tothe Peer Review Group, consisting of, from out-side the Bank, Jane Pratt from the MountainInstitute, John Cool (consultant), and J.D. VonPischke (consultant); and, from inside the Bank,Kathryn McPhail, David Marsden, Anis Dani,and Richard Rosenberg from the ConsultativeGroup to Assist the Poorest (CGAP). WilliamHurlbut and Caroline McEuen edited the report.

Helen Phillip provided administrative supportand assistance with graphics. Oliver Rajakarunaalso provided assistance with graphics. AndrewFoster (consultant, Brown University) providedanalysis of the socio-economic data. The teamconsisted of: Ridley Nelson (team leader), JulianBlackwood (consultant), Maliha Hussein (con-sultant), Sarah Ladbury (consultant), MasaharuShimizu, Roger Slade (consultant), and RashedUl Qayyum. John Heath was also a team mem-ber during the preparation phase.

This report was published as part of theOEDPK Outreach and Dissemination program bya team under the direction of Elizabeth Campbell-Pagé (Task Manager), including Caroline McEuen(Editor) and Juicy Qureishi-Huq (DisseminationCoordinator).

Director-General, Operations Evaluation: Robert Picciotto

Director, Operations Evaluation Department: Gregory K. Ingram

Manager, Sector and Thematic Evaluation: Alain Barbu

Task Manager: Ridley Nelson

v i i

FOREWORD

This is the fourth independ-ent evaluation by the World

Bank’s Operations EvaluationDepartment of the Aga Khan RuralSupport Program (AKRSP) in north-ern Pakistan. Requested and fundedby the Aga Khan Foundation (AKF)and some of its co-donors, the eval-uation was undertaken to assist theAKRSP in shifting its strategy toaccommodate a changing environ-ment and new challenges. It lookedat both the period since the last,1995 evaluation and the full periodsince program initiation in 1982.

Now in its nineteenth year, theprogram has spawned a numberof similar programs both withinPakistan and elsewhere. This eval-uation offers not only the AKFand the AKRSP, but also theglobal rural development commu-nity, including the World Bank, anopportunity to learn from aninnovative and successful pro-gram that now covers about 90percent of the households in thearea it serves.

Some of the lessons from thisexperience have broad relevancefor the rural developmentcommunity:• In rural areas, broad and sus-

tained gains in economic andsocial welfare must be basedon stable increases in output.Increased output requires thatappropriate technology beavailable that yields relativelyquick returns.

• The program has pursued ahigh input/high output strategythat has exploited complemen-

PREFACIO

Ésta es la cuarta evaluaciónindependiente que el Departa-

mento de Evaluación de Operacionesdel Banco Mundial realiza sobre elPrograma Aga Khan de apoyo al sec-tor rural (AKRSP, por su sigla eninglés) en el Pakistán septentrional.La evaluación, solicitada y financiadapor la Fundación Aga Khan y algunosde sus codonantes, se llevó a cabocon el fin de ayudar al AKRSP a aco-modar su estrategia a las nuevas cir-cunstancias y desafíos. Se considerótanto el período posterior a la últimaevaluación, en 1995, como todo eltiempo transcurrido desde la inicia-ción del programa, en 1982.

El programa, que cuenta ya con

19 años, ha generado una serie de

iniciativas semejantes, tanto dentro

de Pakistán como en otros países.

Esta evaluación ofrece, no sólo a la

Fundación Aga Khan y al AKRSP

sino también a las instituciones de

todo el mundo interesadas en el

desarrollo rural, incluido el Banco

Mundial, una oportunidad de

aprender de un programa innova-

dor y fructífero, que abarca ahora

más del 90% de los hogares de la

zona donde se está llevando a

cabo.

Algunas de las enseñanzas de

esta experiencia son muy valiosas

para todos los interesados en el

desarrollo rural:

• En las zonas rurales, los progre-

sos amplios y sostenidos del

bienestar económico y social

deben estar basados en aumen-

tos estables de la producción.

Para aumentar la producción se

requieren tecnologías adecuadas

AVANT-PROPOS

Ce rapport constitue la qua-trième évaluation indépen-

dante du Programme d’appui rural AgaKhan (PARAK) dans le nord du Pakis-tan, réalisée par le Départementd’évaluation des opérations de la Ban-que mondiale. Commissionnée etfinancée par la Fondation Aga Khan(FAK) et certains de ses co-donateurs,l’évaluation a été entreprise dans lebut d’aider le PARAK à réorienter sastratégie afin d’accommoder un envi-ronnement en mutation et de nouvel-les réalités. Il couvre non seulementla période depuis l’évaluation précé-dente de 1995, mais aussi toute lapériode écoulée depuis le lancementdu programme en 1982.

Maintenant dans sa dix-neu-

vième année, le programme donne

naissance à d’autres programmes

similaires tant au Pakistan

qu’ailleurs. La présente évaluation

offre non seulement à la FAK et au

PARAK, mais aussi à toute la com-

munauté de développement rural, y

compris la Banque mondiale,

l’occasion de tirer des leçons d’un

programme novateur réussi qui

couvre aujourd’hui 90 % environ

des ménages dans la région

desservie.

Certaines leçons tirées de cette

expérience présentent un grand

intérêt pour la communauté de

développement rural.

• Dans les zones rurales, les

grands gains concernant le bien-

être économique et social doi-

vent être basés sur un

accroissement stable du rende-

ment. Ce rendement accru exige

qu’une technologie appropriée

FR

AN

ÇA

IS

ES

PA

NO

L

EN

GL

ISH

tarities among the pro-gram components andefficiently used theregion’s scarce commu-nity facilitation and pro-gram management skills.

• Expansion (scaling up)is a slow process even

with skilled facilitators, as thereare few economies of scale.However, expansion by graft-ing new programs into loca-tions with similar circumstancesoffers substantial leverage.

• Partnerships—between gov-ernment, civil society, NGOs,and the private sector—withactionable and measurablecommitments should be for-mally agreed at the start of aprogram to ensure clarity ofroles and to create incentivesfor performance.

• Eventual exit at the communitylevel should be a phasedprocess of increasingly differ-entiated but diminishing com-munity support tailored to, andpartly by, each community,with whom such strategiesshould be agreed up front.

This evaluation was largelycompleted before September 11,2001. Clearly the subsequentevents will have a substantialimpact on the Northern Areas andChitral, which have social andeconomic similarities to neighbor-ing Afghanistan, and some of thefindings here may be relevant tothe reconstruction of Afghanistan.At least temporarily, this regionmay experience increased localsensitivity to sectarian issues,some initial diversion of govern-ment attention from developmentissues to security, and a signifi-cant increase in donor support for

que permitan conseguir

rendimientos relativa-

mente rápidos.

• El programa ha adoptado

una estrategia de altos

insumos y producción

elevada, que ha aprove-

chado las complementa-

riedades entre los componentes

y ha utilizado con eficiencia la

escasa capacidad de facilitación

comunitaria y gestión de progra-

mas de la región.

• La expansión (reproducción en

escala superior) es un proceso

lento aun cuando se disponga de

personal experimentado, ya que

hay pocas economías de escala.

No obstante, la expansión

basada en la introducción de

nuevos programas en lugares

con circunstancias semejantes

puede tener un considerable

efecto multiplicador.

• Las asociaciones —entre el gob-

ierno, la sociedad civil, las ONG y

el sector privado— con compro-

misos aplicables y cuantificables

deberían aprobarse por consenso

al comienzo del programa, para

garantizar la claridad de funciones

e incentivar el desempeño.

• La estrategia de salida debería

ser un proceso gradual de apoyo

comunitario cada vez más dife-

renciado pero decreciente, de

acuerdo con las necesidades de

cada comunidad, con la que

deberían aprobarse inicialmente

dichas medidas.

Esta consulta se terminó en

buena parte antes del 11 de sep-

tiembre de 2001. Obviamente, los

acontecimientos posteriores tendrán

notable repercusión en las zonas

septentrionales y el distrito de Chi-

tral, que presentan semejanzas

sociales y económicas con el vecino

soit disponible et engen-

dre un retour sur l’inves-

tissement relativement

immédiat.

• Le programme a pour-

suivi une stratégie de

hauts intrants/hauts ren-

dements qui exploite les

complémentarités des compo-

sants du programme et utilise

efficacement les rares compéten-

ces de facilitation communautai-

res et de gestion du programme

dans la région.

• L’expansion (augmentation gra-

duelle) est un processus lent

même avec l’aide de facilitateurs

qualifiés, car les économies

d’échelle sont peu nombreuses.

Cependant, l’expansion par

greffe de nouveaux programmes

dans des contextes similaires a

de grands avantages.

• Les partenariats — entre le gou-

vernement, la société civile, les

ONG et le secteur privé —

accompagnés d’engagements

donnant droit d’action et de

mesure devraient être conclus

dès le début d’un programme

afin d’assurer la définition claire

des rôles et de créer des incita-

tions à la performance.

• Le désengagement éventuel au

niveau communautaire devrait

être un processus échelonné de

l’appui communautaire de plus

en plus différencié mais décli-

nant adapté à, et en partie par,

chaque communauté avec laque-

lle ces stratégies ont été conve-

nues au départ.

Le DEO a consulté toutes les

parties prenantes clés et a visité un

village échantillon. Les points de

vue, y compris les avis dissidents,

ont été minutieusement considérés.

Les opinions exprimées dans le pré-

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

v i i i

EN

GL

ISH

FR

AN

ÇA

IS

ES

PA

NO

L

Pakistan. This does notsubstantially change thedirection of the findingsand recommendations, butthe climate of uncertaintymay call for greater flexibil-ity regarding the rate of theshift. This does not neces-

sarily mean a slow-down. Oppor-tunities may be created thatshould be seized quickly, and itwill be essential to remain intouch with the needs of theclients in the villages, particularlythe needs of the various groupswithin villages and within com-munity organizations.

OED consulted with all the keystakeholders and visited a sampleof villages. Their views, includingthose in dissent, were carefullyconsidered. The opinionsexpressed in this report do notnecessarily represent the views ofthe AKF, the AKRSP, the otherdonor agencies, the governmentof Pakistan, or the World Bank.

Afganistán, y algunas de las

conclusiones aquí ofrecidas

pueden ser de interés para

la reconstrucción de Afganis-

tán. Al menos temporal-

mente, esta región puede

experimentar una mayor

sensibilidad local a las cues-

tiones sectarias, cierto abandono

gubernamental de las cuestiones

del desarrollo en aras de la seguri-

dad, y un aumento significativo del

apoyo de los donantes a Pakistán.

Ello no cambia sustancialmente la

dirección de las conclusiones y

recomendaciones, pero el clima de

incertidumbre puede requerir

mayor flexibilidad sobre el ritmo

del cambio. Todo esto no significa

necesariamente una desaceleración.

Pueden surgir oportunidades que

deberían aprovecharse de inme-

diato, y será fundamental permane-

cer en contacto con las necesidades

de los clientes en las aldeas, en par-

ticular con las necesidades de los

distintos grupos dentro de las

aldeas y dentro de la organización

comunitaria.

El DEO ha consultado a todas

las principales partes interesadas y

ha visitado una muestra de aldeas.

Se tuvieron muy en cuenta sus opi-

niones, aun cuando se mostraron

en desacuerdo. Las opiniones reco-

gidas en este informe no represen-

tan necesariamente las de la

Fundación Aga Khan, el AKRSP, los

otros organismos donantes, el gob-

ierno de Pakistán o el Banco

Mundial.

sent rapport ne reflètent pas

nécessairement les points de

vue de la FAK, du PARAK ou

d’autres organismes de

donateurs, du gouvernement

pakistanais ou de la Banque

mondiale.

F o r e w o r d

i x

EN

GL

ISH

FR

AN

ÇA

IS

ES

PA

NO

L

Robert PicciottoDirector-General, Operations Evaluation

x i

EXECUTIVESUMMARY

The Aga Khan Rural SupportProgram (AKRSP), like many

a traveler before them in these highmountains, has reached a summit,only to see a greater peak ahead.For 18 years the AKRSP has helpedcommunity groups throughout theNorthern Areas and Chitral Districtof Pakistan in a development effortthat has become a model for ruralprograms throughout the countryand across the globe. But donorfatigue is now raising the prospectof a leaner future, community andarea development is becoming morecomplex, maintaining incentives forparticipation faces challenges, andthe poorer areas and people theAKRSP is now focusing on needeven greater creative effort toachieve success. As a new donorfunding cycle approaches, theAKRSP is gauging the path of itsnext ascent, taking stock of its manyaccomplishments, and seeking todefine its future relationship with allof its stakeholders. This evaluation,the fourth commissioned since 1986from the Operations EvaluationDepartment (OED) of the WorldBank, is intended to assist thatprocess.

The evaluation assesses thedevelopment outcome of both thefull 18-year life of the AKRSP andthe 5-year period since the lastevaluation. It focuses on four pro-gram components: communityorganizations, infrastructuredevelopment, natural resourcemanagement, and microfinance.The criteria that have been usedto assess program performance

RESUMEN

El Programa Aga Khan deapoyo al sector rural (AKRSP,

por su sigla en inglés), como muchosotros viajeros que le precedieron enestas altas montañas, ha coronadouna cima, para encontrarse con unacumbre todavía más alta. Desde hace18 años, el AKRSP ayuda a los gruposcomunitarios de todas las zonas sep-tentrionales y del distrito de Chitral(ZSC), en Pakistán, en una iniciativade desarrollo que se ha convertido enmodelo para programas rurales en elpaís y en todo el mundo. Pero la fatigade los donantes está llevando ahora aplantearse la perspectiva de un futuromenos ambicioso, el desarrollo de lascomunidades y regiones está adqui-riendo mayor complejidad, el manteni-miento de incentivos para laparticipación se encuentra con impor-tantes problemas y las zonas y perso-nas más pobres en que el AKRSP estáconcentrando ahora su atención nece-sitan un esfuerzo todavía más crea-tivo. Al acercarse un nuevo ciclo definanciamiento de los donantes, elAKRSP está tratando de determinar elitinerario de su próxima escalada,tomar nota de sus numerosos logros yaclarar su relación futura con todaslas partes interesadas. Esta evalua-ción, la cuarta solicitada desde 1986al Departamento de Evaluación deOperaciones (DEO) del Banco Mun-dial, tiene como objetivo contribuir aese proceso.

En esta evaluación se determi-

nan los resultados en términos de

desarrollo tanto de los 18 años de

vida del AKRSP como de los cinco

años transcurridos desde la última

evaluación. Se presta especial aten-

RÉSUMÉ ANALYTIQUE

Le Programme d’appui ruralAga Khan (PARAK), comme

tant d’autres avant lui dans ces chaî-nes montagneuses, n’a atteint un som-met que pour en apercevoir un autreplus grand devant lui. Depuis 18 ans,le PARAK aide les groupements com-munautaires des zones du Nord et dudistrict Chitral au Pakistan dans leureffort de développement qui estdevenu un exemple de programmerural dans tout le pays et dans lemonde entier. Mais la fatigue desorganismes donateurs laisse à penserque l’avenir sera plus difficile ; eneffet, le développement zonal et com-munautaire devient de plus en pluscomplexe, et le maintien des incita-tions à la participation pose des défis ;ainsi, les zones pauvres et populationsciblées par le PARAK devront bénéfi-cier de plus grands efforts novateurspour réussir. Alors que les donateursentrent dans un nouveau cycle definancement, le PARAK explore lechemin à suivre pour sa prochaineascension, tout en faisant le bilan deses nombreuses activités et en cher-chant à définir ses futures relationsavec toutes ses parties prenantes. Laprésente évaluation (la quatrièmecommissionnée depuis 1986 par leDépartement de l’évaluation des opé-rations (DEO) de la Banque mondiale,a pour but de faciliter ce processus.

L’évaluation fait le bilan des

résultats du développement à la fois

pendant les 18 années de durée de

vie du PARAK et la période de 5

ans qui s’est écoulée depuis l’éva-

luation précédente. Elle est centrée

sur quatre composantes du pro-

gramme : les organisations commu-

FR

AN

ÇA

IS

ES

PA

NO

L

EN

GL

ISH

are those used by OED forWorld Bank projects: rele-vance, efficacy, efficiency,institutional developmentimpact, and sustainability.

The methods used inthis evaluation consistedmainly of group and indi-

vidual discussions in a sample of24 villages selected to represent aspread of village types and capac-ities. It also included economicanalysis; a microfinance staffworkshop; application of foursemiformal village questionnaires;an institutional survey of AKRSPstaff; analysis of socio-economicsurvey data collected in 1991 and1997; review of the extensiveAKRSP literature, including otherdonor evaluations; an investiga-tion of cost comparators of otherrural development projects; anddiscussions with senior staff of theprogram components and theadministrations.

To assess the challenges ofthis program it is necessary tounderstand that the NorthernAreas (although not Chitral,which is part of North West Fron-tier Province) is militarily sensitiveand politically different from therest of Pakistan. For historicalreasons, it is a federally adminis-tered area, under the jurisdictionof Pakistan’s Federal Minister ofKashmir Affairs and NorthernAreas and a Legislative Council,headed by a federally appointedchief executive and consisting of24 locally elected members whoelect a deputy chief executive.The people of the NorthernAreas do not elect members toPakistan’s National Assembly,nor are they directly taxed, butthey are provided a number ofsubsidies.

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

x i i

EN

GL

ISH

ción a cuatro componentes

del programa: organizacio-

nes comunitarias, desarrollo

de la infraestructura, gestión

de los recursos naturales y

microfinanciamiento. Los cri-

terios utilizados para evaluar

el desempeño del programa

son los empleados por el DEO para

los proyectos del Banco Mundial:

pertinencia, eficacia, eficiencia,

efectos en el desarrollo institucional

y sostenibilidad.

Los métodos empleados para

esta evaluación han sido sobre todo

conversaciones colectivas e indivi-

duales en una muestra de 24 aldeas

seleccionadas en representación de

una gran diversidad de tipos de

aldeas y capacidades. Se han utili-

zado también los siguientes medios:

el análisis económico, un taller

sobre microfinanciamiento para el

personal, la aplicación de cuatro

cuestionarios semiformales en las

aldeas, una encuesta institucional

del personal del AKRSP, el análisis

de datos socioeconómicos recopila-

dos mediante encuestas entre 1991

y 1997, el estudio de las numerosas

publicaciones del AKRSP (incluidas

las evaluaciones de otros donantes),

una investigación de los compara-

dores de costos de otros proyectos

de desarrollo rural y conversaciones

con el personal de rango superior

sobre los componentes del pro-

grama y las administraciones.

Para evaluar los desafíos que

presenta este programa es preciso

comprender que las zonas septen-

trionales (aunque no Chitral, que

forma parte de la Provincia de la

Frontera Noroccidental) son de

importancia crítica desde el punto

de vista militar y políticamente dife-

rentes del resto de Pakistán. Por

razones históricas, se trata de una

zona administrada federalmente,

nautaires, le développement

de l’infrastructure, la gestion

des ressources naturelles et

le microfinancement. Les cri-

tères utilisés pour évaluer la

performance du programme

sont ceux que le DEO utilise

pour les projets de la Ban-

que mondiale : pertinence, effica-

cité, impact sur le développement

institutionnel et durabilité.

Les méthodes utilisées pour

l’évaluation ont été principalement

des discussions en groupes et indi-

viduelles dans 24 villages-échan-

tillons sélectionnés de façon à

représenter la diversité des villages

et de leurs capacités. Elles ont éga-

lement consisté en des analyses

économiques, un atelier pour le

personnel sur le microfinancement,

l’application de quatre questionnai-

res villageois semi-formels, une

enquête institutionnelle du person-

nel du PARAK, l’analyse des don-

nées de l’enquête

socio-économique recueillies en

1991 et 1997, une revue de la vaste

litérature sur le PARAK (y compris

les évaluations effectuées par

d’autres organismes donateurs), une

enquête sur les comparateurs de

coûts d’autres projets de développe-

ment rural, et des discussions avec

les cadres responsables des compo-

santes du programme et des admin-

istrations.

Avant d’aborder les défis de ce

programme, il est nécessaire de

comprendre que les zones du nord

(sauf la zone Chitral qui est située

dans la Province frontière nord-

ouest) constituent une région criti-

que en termes militaires et

politiquement différente du reste du

pays. Pour des raisons historiques,

c’est une zone administrée par le

gouvernement fédéral pakistanais

sous la juridiction du ministre fédé-

ES

PA

NO

L

FR

AN

ÇA

IS

OutcomeThe achievements of theAKRSP have been highlysatisfactory. Results havebeen remarkable over bothits 18-year life and its mostrecent 5-year period. Whileweaknesses have appeared

in some areas, they are matchedby strengths rarely found in ruraldevelopment interventions. Fur-thermore, based on the program’squick and creative responses toemerging problems in the past,the AKRSP can be expected tomodify its strategy and tactics asnew priorities and opportunitiesappear.

By adopting a flexibleapproach and learning fromexperience, the program hasmaintained substantial relevanceto the development priorities ofPakistan and the Northern Areasand Chitral (NAC) since its incep-tion in late 1982. More recently,however, its relevance has beenthreatened by continued weak-ness in government capacity;declining frequency of infrastruc-ture investment at the villagelevel that in the past served tobind community organizations ina common purpose; increasedpluralism in community organiza-tions; too little progress in bring-ing women into the mainstream;declines in saving and creditflows; and increased overdues inmicrofinance. These changes inthe program environment, whilechallenging, also present renewaland partnership opportunities forthe AKRSP.

Efficacy (achievement of objec-tives) has been fully satisfactory,in most respects highly satisfac-tory. Incomes have increased sub-stantially, certainly beyond the

E x e c u t i v e S u m m a r y

x i i i

EN

GL

ISH

bajo la jurisdicción del

Ministro Federal de Asuntos

de Cachemira y las Zonas

Septentrionales y un Consejo

Legislativo, presidido por un

jefe ejecutivo de nombra-

miento federal e integrado

por 24 miembros localmente

elegidos, que eligen a su vez a un

jefe ejecutivo adjunto. La población

de las zonas septentrionales no

elige a los miembros de la Asam-

blea Nacional de Pakistán ni está

sometida a impuestos directos, pero

recibe algunos subsidios.

ResultadoLos logros del AKRSP han sido muy

satisfactorios. Los resultados han

sido notables tanto en sus 18 años

de vida como en los cinco últimos

años. Si bien se han detectado defi-

ciencias en algunas esferas, están

contrarrestadas por ventajas que

rara vez se encuentran en interven-

ciones en favor del desarrollo rural.

Además, habida cuenta de las rápi-

das y creativas respuestas del pro-

grama a los problemas aparecidos

en el pasado, es de prever que el

AKRSP modifique su estrategia y

tácticas a medida que aparezcan

nuevas prioridades y oportunidades.

Al adoptar un planteamiento flexi-

ble y aprender de la experiencia, el

programa ha mantenido una perti-

nencia sustancial en relación con las

prioridades de desarrollo de Pakistán

y las zonas septentrionales y Chitral

desde su comienzo, a finales de

1982. No obstante, más reciente-

mente su pertinencia se ha visto

amenazada por las persistentes defi-

ciencias de la capacidad de gob-

ierno; la menor frecuencia de

inversiones en infraestructura en las

aldeas, que en el pasado contribuye-

ron a unir a las organizaciones

comunitarias en la búsqueda de un

ral des Affaires du Cache-

mire et des zones septentrio-

nales, et d’un Conseil

législatif dirigé par un pre-

mier magistrat désigné au

plan fédéral, qui comprend

24 membres élus localement

et chargés d’élire un sous-

premier magistrat. Les populations

des zones du nord n’élisent pas les

membres de l’Assemblée nationale

pakistanaise et ne sont pas soumi-

ses à l’impôt direct, mais elle béné-

ficient de subventions.

RésultatsLes résultats du PARAK sont très

satisfaisants. Ils ont été remarqua-

bles aussi bien pendant ses 18 ans

d’exécution que pendant sa période

plus récente de 5 ans. Si on note

des faiblesses dans certains domai-

nes, elles sont amplement compen-

sées par des forces rarement

relevées dans les interventions de

développement rural. De plus, si

l’on se base sur la réaction rapide et

novatrice du programme face aux

problèmes émergents du passé, on

peut anticiper que le PARAK sera

en mesure de modifier sa stratégie

et sa tactique à mesure qu’apparaî-

tront de nouvelles priorités et

opportunités.

En adoptant une démarche sou-

ple et en tirant des leçons de

l’expérience, le programme est

demeuré en grande partie pertinent

pour ce qui est des priorités du

Pakistan et des Zones du nord/Chi-

tral (ZNC) depuis son lancement en

fin d’année 1982. Toutefois, sa per-

tinence a été mise en danger

récemment en raison de la médio-

cre performance persistante du

gouvernement, du déclin de la fré-

quence des investissements dans

l’infrastructure villageoise qui autre-

fois liaient les organisations com-

ES

PA

NO

L

FR

AN

ÇA

IS

original target of a dou-bling in real terms. It is dif-ficult to prove attribution atthe aggregate level absenta “without program” con-trol, but both the economicanalysis and analysis of the1991 and 1997 socio-eco-

nomic data suggest that a share ofthese benefits—more than suffi-cient to justify the cost—is attrib-utable to the AKRSP. The programappears to have been very effec-tive in enhancing beneficiarycapacity and building social capi-tal. The objective of replicabilityhas been fully achieved, bothwithin Pakistan and elsewhere,and in many respects the programhas become a laboratory for ruraldevelopment. About eight majorprograms or projects in Pakistanhave drawn substantially fromAKRSP experience, including theNational Rural Support Programand the ongoing World Bank–sup-ported Poverty Alleviation Project.Outside Pakistan, other pro-grams—supported not only by theAga Khan Foundation but byother donors as well—use com-munity-based processes drawnsubstantially or partly from theAKRSP. While efficacy in achiev-ing the AKRSP’s earlier objectiveof “working itself out of a job”has been negligible, it is question-able whether this was a realisticobjective at the time. Even todayit is probably unrealistic at a pro-gram level, although it is realisticfor an increasing number of moremature community organizations.

Efficiency has been satisfactoryin terms of the costs of achievingthese results. While the AKRSP isat the top end of a range ofglobal and local comparators foroperating costs per household,

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

x i v

EN

GL

ISH

objetivo común; el mayor

pluralismo de las organizacio-

nes comunitarias; el escaso

progreso en la integración de

la mujer; el descenso de los

flujos de ahorro y crédito, y

el aumento de las deudas

pendientes en las actividades

de microfinanciamiento. Estos cam-

bios en el entorno del programa, si

bien representan un desafío, ofrecen

también oportunidades de renova-

ción y asociación para el AKRSP.

La eficacia (logro de los objeti-

vos) ha sido satisfactoria y, en la

mayor parte de los aspectos, muy

satisfactoria. Los ingresos han

aumentado de forma sustancial, cier-

tamente más del objetivo original de

duplicarlos en términos reales. Es

difícil determinar la atribución en

cifras globales, dada la ausencia de

una hipótesis de control “sin pro-

grama”, pero tanto el análisis econó-

mico como el estudio de los datos

socioeconómicos de 1991 y 1997

parecen indicar que una parte de

estos beneficios —más de lo sufi-

ciente para justificar el costo— es

atribuible al AKRSP. Parece que el

programa ha sido muy eficaz en lo

que respecta a aumentar la capaci-

dad de los beneficiarios e incremen-

tar el patrimonio social. El objetivo

de posibilidad de reproducción se

ha conseguido plenamente, tanto

dentro de Pakistán como en otros

lugares, y en muchos sentidos el

programa se ha convertido en un

laboratorio para el desarrollo rural.

Unos ocho grandes programas o

proyectos de Pakistán se han basado

ampliamente en la experiencia del

AKRSP, incluido el programa nacio-

nal de apoyo rural y el actual pro-

yecto de reducción de la pobreza

respaldado por el Banco Mundial.

Fuera de Pakistán, otros programas

—patrocinados no sólo por la Fun-

munautaires à leur but com-

mun, le pluralisme croissant

des organisations commu-

nautaires, le peu de progrès

vers l’intégration des femmes

dans le processus général, le

déclin de l’épargne et de cir-

culation du crédit, et les

arriérés de paiement du microfinan-

cement de plus en plus nombreux.

Bien que redoutables, ces change-

ments dans l’environnement du

programme sont aussi pour le

PARAK des opportunités de renou-

veau et de partenariats.

L’efficacité (atteinte des objectifs)

est très satisfaisante à bien des

égards. Les revenus ont augmenté

sensiblement, certes bien au-delà

de la cible initiale de doublement

en termes réels. Il est difficile d’en

prouver l’attribution au niveau

d’agrégat en l’absence de contrôle

« sans programme » ; mais l’analyse

économique et celle des données

socio-économiques effectuées en

1991 et 1997 montrent qu’une partie

de ces avantages — plus que suffi-

sante pour en justifier le coût — est

imputable au PARAK. Le pro-

gramme semble avoir été très effi-

cace en matière d’amélioration de la

capacité des bénéficiaires et d’aug-

mentation du capital social. L’objec-

tif de duplication a été pleinement

atteint aussi bien au Pakistan

qu’ailleurs et, à bien des égards, le

programme est devenu un labora-

toire de développement rural. Au

Pakistan, environ huit grands pro-

jets ont tiré des leçons de l’expé-

rience du PARAK, dont le

Programme national d’appui rural et

le Projet de réduction de la pau-

vreté en cours cofinancé par la

Banque mondiale. En dehors du

Pakistan, d’autres programmes —

cofinancés non seulement par la

Fondation Aga Khan mais aussi par

ES

PA

NO

L

FR

AN

ÇA

IS

the estimated economicrate of return of at least 16percent—probably closerto 25 to 30 percent if allbenefits could be quanti-fied and attributed—sug-gests the investmentchoices have been sound.

It should be possible, however, toimprove efficiency even furtherthrough greater tailoring of com-munity support to communityself-help capacity, throughincreased collaboration within theAga Khan Development Network(AKDN) family, and through ahigher level of cost recovery forinvestments, particularly for pri-vate goods and for training.

Institutional DevelopmentImpactCommunity organizations lie atthe heart of the AKRSP approach.Village organizations started bythe program have been the mech-anism for developing social andhuman capital; creating infrastruc-ture; carrying out savings andloan activities; improving agricul-ture, livestock, and forestry; andproviding a convenient channelfor government and other agen-cies to respond to village needs.The achievements have beenimpressive and, unlike manyother donor-funded interventions,sustained. Currently, two maintypes of organizations are beingsupported: village organizationsfor men and women’s organiza-tions, although umbrella localdevelopment organizations arebecoming increasingly important.Within the villages there is wide-spread acknowledgment of whatthese organizational structureshave done for members, andthere is survey evidence that

E x e c u t i v e S u m m a r y

x v

EN

GL

ISH

dación Aga Khan sino tam-

bién por otros donantes—

utilizan procesos de base

comunitaria inspirados en

mayor o menor medida en el

AKRSP. Si bien la eficacia en

el logro del objetivo inicial

de “llegar a no ser necesa-

rios” ha sido insignificante, quizá

este objetivo no era muy realista en

aquellas fechas. Incluso hoy día es

probablemente poco realista desde

la perspectiva del programa, aunque

puede serlo para un número cre-

ciente de organizaciones comunita-

rias más maduras.

La eficiencia ha sido satisfactoria

en lo que respecta a los costos con-

traídos para alcanzar esos resulta-

dos. Si bien el AKRSP se encuentra

en el extremo superior de una serie

de comparadores mundiales y loca-

les en lo que se refiere a los costos

de explotación por hogar, la tasa de

rentabilidad económica estimada de

al menos el 16% —y probablemente

más próxima al 25%-30%, si se

cuantifican y atribuyen todos los

beneficios— parece indicar que las

decisiones de inversión han sido

acertadas. No obstante, sería posible

lograr una eficiencia todavía mayor

mediante una mejor adaptación del

apoyo comunitario a la capacidad

de autoayuda de la comunidad, una

mayor colaboración con la Red de

Desarrollo Aga Khan (AKDN, por

sus siglas en inglés) y un mayor

nivel de recuperación de costos

para las inversiones, sobre todo en

lo que se refiere a los bienes priva-

dos y la capacitación.

Efectos en el desarrolloinstitucionalLas organizaciones comunitarias

ocupan un lugar central en el plan-

teamiento del AKRSP. Las organiza-

ciones rurales iniciadas por el

d’autres organismes dona-

teurs — utilisent des proces-

sus communautaires tirés en

tout ou partie du PARAK.

Bien que l’efficacité pour

atteindre l’objectif établi plus

tôt et qui consistait à « ne

plus rien avoir à faire » ait

été négligeable, il est maintenant

présumé douteux qu’il ait été réal-

iste à l’époque. Même aujourd’hui,

il est probablement chimérique au

niveau d’un programme, mais il

reste réaliste au niveau d’un nom-

bre croissant d’organisations com-

munautaires plus matures.

L’efficacité a été atteinte en ter-

mes de coûts pour arriver à ces

résultats. Bien que le PARAK se

place à l’extrémité supérieure d’une

plage de comparateurs mondiaux et

locaux des coûts de fonctionnement

par ménage, le taux de rentabilité

économique estimé à 16 pour cent

au moins — probablement plus

près de 25 à 30 pour cent si tous

les avantages pouvaient être quanti-

fiés et imputés — laisse à penser

que les investissements choisis

étaient solides. Il devrait toutefois

être possible d’améliorer l’efficacité

encore davantage à travers l’adapta-

tion plus poussée de l’appui com-

munautaire à la capacité d’entraide

des communautés, grâce à une col-

laboration plus serrée au sein du

groupe de Réseaux de développe-

ment Aga Khan (RDAK) et d’un

plus haut niveau de recouvrement

des coûts d’investissement, en parti-

culier dans les biens privés et la for-

mation professionnelle.

Impact sur le développementinstitutionnelLes organisations communautaires

sont au cœur de l’approche du

PARAK. Les organisations villageoi-

ses établies par le programme ont

ES

PA

NO

L

FR

AN

ÇA

IS

being in a village withcommunity organizationsbrings a range of benefits.

Based on the strength ofthe program’s communityorientation, these organiza-tions, or their evolvingforms, will likely remain

the locus of the AKRSP’s develop-ment effort. In the future, how-ever, resource constraints andsustainability goals will dictate anincreasing degree of differentia-tion in the effort the AKRSPdevotes to different types oforganizations. For example, theAKRSP may have to choosebetween supporting high-flierswith good business skills relevantto the needs of the local econ-omy; or organizations that haveprimarily social objectives, such asmanaging conflict resolution orcanvassing for girls’ education; orsimply the most economically dis-advantaged areas and villages.Designing an effective, differenti-ated approach will call for a goodunderstanding of the differencesin village institutional maturity,the reasons for those differ-ences—the status of social capital,the needs of different groups—and for evaluation of the relativepoverty impacts of alternativeapproaches.

The main weakness of theAKRSP, which owes its originpartly to the very success of theprogram, is related to institutionaldevelopment and institutional sus-tainability in the broadest sense. Ifthe AKRSP were to close tomor-row, there would be a large insti-tutional gap in the NAC, a gapthat other agencies, most signifi-cantly the district administrations,could not fill. While the AKRSPhas contributed positively to gov-

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

x v i

EN

GL

ISH

programa han sido el meca-

nismo utilizado para el des-

arrollo del capital social y

humano, la creación de

infraestructura, la realización

de actividades de ahorro y

préstamo, la mejora de la

agricultura, la ganadería y la

silvicultura, y el establecimiento de

un cauce válido para que el gob-

ierno y otros organismos respondan

a las necesidades de las aldeas. Los

logros han sido impresionantes y, a

diferencia de muchas otras inversio-

nes financiadas por donantes, dura-

deros. En la actualidad, se presta

apoyo a dos tipos principales de

organización —organizaciones rura-

les para hombres y organizaciones

de mujeres—, aunque cada vez

están adquiriendo mayor importan-

cia las organizaciones de desarrollo

rural de carácter más amplio. Dentro

de las aldeas se reconoce en general

lo que estas estructuras organizativas

han hecho por sus miembros, y en

las encuestas hay testimonios de que

el pertenecer a una aldea con orga-

nizaciones comunitarias representa

una serie de beneficios.

Teniendo en cuenta la importan-

cia de la orientación comunitaria del

programa, estas organizaciones, o las

que resulten de su evolución, conti-

nuarán siendo probablemente el

núcleo de los esfuerzos de desarrollo

del AKRSP. No obstante, en el futuro

las limitaciones de recursos y los

objetivos de sostenibilidad impon-

drán un grado cada vez mayor de

diferenciación en el esfuerzo que el

AKRSP dedicará a los diferentes tipos

de organización. Por ejemplo, es

posible que el AKRSP tenga que ele-

gir entre apoyar a las personas más

ambiciosas con dotes empresariales

valiosas para la economía local o a

las organizaciones que tienen funda-

mentalmente objetivos sociales,

servi de mécanismes de

développement du capital

social et humain en créant

l’infrastructure, en réalisant

des activités d’épargne

immobilière, en améliorant

les secteurs de l’agriculture,

de l’élevage et de la foreste-

rie, et en servant de réseau à la dis-

position des gouvernements et

autres organismes donateurs pour

répondre aux besoins villageois. Les

résultats ont été impressionnants et,

contrairement à d’autres interven-

tions financées par des donateurs,

ils sont durables. Pour l’instant,

deux grands types d’organisations

bénéficient d’un appui : les organi-

sations villageoises masculines et

les organisations villageoises fémini-

nes dans un milieu où les organis-

mes ombrelles de développement

rural deviennent de plus en plus

importants. Dans les villages, on

sait très bien ce que ces structures

organisationnelles font pour leurs

membres et, d’après les enquêtes, il

est évident que le fait d’appartenir à

un village doté d’organisations com-

munautaires apporte toutes sortes

d’avantages.

En raison de l’orientation com-

munautaire solide du programme, il

est probable que ces organisations

(ou leurs formes évoluées) demeu-

reront le locus de l’effort de déve-

loppement du PARAK. Dans l’avenir

cependant, les contraintes de res-

sources et la durabilité des résultats

seront les préceptes du degré crois-

sant de différentiation de l’effort

consacré par le PARAK aux divers

types d’organisations. Par exemple,

le PARAK devra peut-être choisir

entre l’appui aux « élites » possédant

une expertise commerciale solide

pour satisfaire les besoins de l’éco-

nomie locale, l’appui à des organi-

sations dont les objectifs seront

ES

PA

NO

L

FR

AN

ÇA

IS

ernment initiatives in theNAC, and even sought toinfluence national policy, itis more difficult to discernthe AKRSP’s handiwork inthe line departments ofgovernment, the privatesector, and provincial and

national policymaking forums. Yetit is success in these arenas thatwill largely determine the courseof the region’s future progress.Without a marked improvementin the government’s developmentcapacity, long-run sustainedprogress in the NAC will remainelusive. While government per-formance is not the AKRSP’sresponsibility, a more equal part-nership of all key players in theNAC should now be seen by theAKRSP, by AKRSP donors, by gov-ernment, and by governmentdonors as the essential locus offuture development and the keyinstrument in best serving thefuture needs of the people of theNAC.

All in all, the AKRSP was foundto be well managed—based partlyon an institutional survey under-taken by the evaluation team.Some shortcomings in theAKRSP’s human resource policiesremain, however, especially thelong-standing issue of genderimbalance. The AKRSP is rightlyplanning to make a number oforganizational changes to betteralign itself to changes in its strate-gic focus. In reorganizing, theAKRSP should continue to beguided by organizational equity asin the past, but it should also beguided by the demands of theoverarching goal of forging adevelopment coalition in the NACthrough interlocking partnershipswith all key players. Reorganizing

E x e c u t i v e S u m m a r y

x v i i

EN

GL

ISH

como lograr la resolución de

conflictos o promover la edu-

cación de las niñas, o simple-

mente a las zonas y aldeas

más desfavorecidas económi-

camente. La designación de

un planteamiento eficaz y

diferenciado requerirá una

comprensión adecuada de las dife-

rencias en la madurez institucional

de las aldeas, las razones de esas

diferencias —situación del patrimo-

nio social, necesidades de los dife-

rentes grupos— y una evaluación de

los aspectos relativos que los distin-

tos enfoques pueden tener en la

pobreza.

La principal deficiencia del

AKRSP, que debe su origen en parte

al éxito mismo del programa, está

relacionada con el desarrollo insti-

tucional y la sostenibilidad institu-

cional en sentido más amplio. Si el

AKRSP desapareciera mañana,

habría una gran brecha institucional

en las ZSC, brecha que otros orga-

nismos, en particular las administra-

ciones de distrito, no podrían

cubrir. Si bien el AKRSP ha contri-

buido positivamente a las iniciativas

gubernamentales en las ZSC, e

incluso trató de influir en las políti-

cas nacionales, es más difícil deter-

minar la labor del AKRSP en los

departamentos sectoriales del gob-

ierno, el sector privado y los foros

de formulación de políticas nacio-

nales y provinciales. No obstante,

es precisamente su éxito en estas

esferas lo que determinará en

buena medida el curso del progreso

futuro de la región. Sin una notable

mejoría de la capacidad de desarro-

llo del gobierno, será imposible

conseguir un progreso sostenido a

largo plazo en las ZSC. Si bien el

desempeño gubernamental no es

responsabilidad del AKRSP, una

asociación más igualitaria de todas

principalement d’ordre

social, tels que la gestion de

la résolution des conflits ou

le démarchage pour l’éduca-

tion des filles, et l’appui sim-

plement aux zones et

villages les plus démunis sur

le plan économique. L’élabo-

ration d’une approche différenciée

efficace exigera une bonne compré-

hension des différences de maturité

des institutions villageoises, les rai-

sons de ces différences — le statut

du capital social, les besoins des

divers groupes — et une évaluation

des impacts relatifs d’autres

approches.

La principale faiblesse du

PARAK, qui est en partie imputable

à la réussite même du programme,

se rapporte au développement et à

la durabilité institutionnels dans le

sens le plus large. Si le PARAK

devait être abandonné demain, il y

aurait un « trou » institutionnel dans

les ZNC que les autres organismes

ne pourraient pas combler, surtout

les administrations au niveau des

districts. Il est certes évident que le

PARAK a contribué de manière

positive aux initiatives gouverne-

mentales dans les ZNC, et a même

recherché à influencer la politique

nationale, mais il est plus difficile

de discerner les activités du PARAK

dans les ministères concernés, le

secteur privé et les forums natio-

naux et provinciaux de prise de

décisions. Et pourtant, c’est la réus-

site dans cette arène qui détermi-

nera en grande partie le cours de

l’évolution future de la région. Sans

amélioration marquée de la capa-

cité de développement gouverne-

mentale, les avancées durables à

long terme dans les ZNC demeure-

ront élusives. Bien que le PARAK

ne soit pas responsable de la per-

formance gouvernementale, un par-

ES

PA

NO

L

FR

AN

ÇA

IS

for partnership will call forarrangements that encour-age the closest possibleinteraction with the keyinstitutions in the NAC,especially with governmentand the AKDN. It is thisgoal that should be given

the greatest weight in framing theAKRSP’s future organization.

SustainabilityThe approach and developmentideas of the AKRSP are sustain-able, as demonstrated by theirreplication elsewhere. Many ofthe village organizations createdby the program have matured andwould likely survive without closeAKRSP supervision—“it would bedifficult, but we would manage,”in the words of one communityleader. The infrastructure builtunder the program is also sustain-able because it was village-chosen, is well maintained, andprovides a positive return. Sus-tainability of the microfinanceprogram has been and remainsgood, despite recent operationalfailures, and the program is nowbeing passed on to a new,AKRSP-controlled bank. Thereare, however, some notable sus-tainability risks.

The most obvious, but onethat is shared with all non-endowment, donor-supportedprograms, is that the AKRSP is notfinancially sustainable. It relies onthe continuing patronage ofdonors. Signs of donor fatiguemay lead to reduced funding inthe future (although recent eventsin the region may change that). Ifthe AKRSP tightens its belt; if itvigorously pursues the goal of adevelopment coalition; if it devel-ops a new strategy pursuing effi-

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

x v i i i

EN

GL

ISH

las principales partes intere-

sadas en las ZSC debería ser

considerada ahora por el

AKRSP, por los donantes del

AKRSP, por el gobierno y

por los donantes del gob-

ierno como elemento funda-

mental del desarrollo futuro

y como instrumento clave para

atender mejor las necesidades futu-

ras de la población de las ZSC.

En conjunto, se ha observado que

el AKRSP está bien administrado,

según resultados basados en parte

en una encuesta institucional reali-

zada por el equipo de evaluación.

No obstante, continúan algunos de

los inconvenientes en las políticas de

recursos humanos del AKRSP, en

particular el problema ya tradicional

del desequilibrio entre el hombre y

la mujer. El AKRSP está planificando

la introducción de una serie de cam-

bios organizativos para acomodarse

mejor a las transformaciones de su

orientación estratégica. En esa reor-

ganización, el AKRSP debería seguir

teniendo como norma la equidad, lo

mismo que en el pasado, pero debe-

ría también tener en cuenta las

demandas del objetivo global de

establecer una coalición para el des-

arrollo en las ZSC mediante asocia-

ciones mutuas con todos los

participantes principales. Esta reorga-

nización exigirá el establecimiento

de mecanismos que alienten la inter-

acción más estrecha posible con las

instituciones clave de las ZSC, espe-

cialmente con el gobierno y la

AKDN. Este objetivo debería recibir la

máxima importancia al configurar

la organización futura del AKRSP.

SostenibilidadEl planteamiento y las ideas de des-

arrollo del AKRSP son sostenibles,

como demuestra su aplicación en

otros lugares. Muchas de las organi-

tenariat mieux équilibré

entre tous les acteurs clés

des ZNC devrait être mainte-

nant considéré par le PARAK

et ses organismes donateurs,

et le gouvernement et ses

organismes donateurs,

comme le locus essentiel du

développement à venir et l’instru-

ment clé pour mieux répondre aux

besoins futurs des populations des

ZNC.

Somme toute, il a été déterminé

que le PARAK avait été bien exé-

cuté — selon, en partie, une

enquête institutionnelle effectuée

par l’équipe d’évaluation. On note

encore cependant quelques insuffi-

sances dans les politiques de res-

sources humaines, notamment la

question ancienne déjà portant sur

l’inégalité entre les sexes. Le PARAK

prévoit à juste titre d’apporter quel-

ques changements organisationnels

afin d’être dans le droit fil de

l’objectif de sa réorientation straté-

gique. Pendant sa réorganisation, le

PARAK devra encore être guidé par

l’équité organisationnelle, mais

aussi par les exigences liées au but

déterminant de forger une coalition

de développement dans les ZNC à

travers des partenariats réunissant

tous les acteurs clés. La réorganisa-

tion en partenariats exigera des

mesures d’encouragement à l’inter-

action la plus étroite possible avec

les institutions clés dans les ZNC,

notamment avec le gouvernement

et le groupe RDAK. C’est ce but qui

devrait avoir le plus de poids dans

l’élaboration de l’organisation

future du PARAK.

DurabilitéL’approche et les idées de dévelop-

pement du PARAK sont durables,

comme l’a démontré sa duplication

ailleurs. De nombreuses organisa-

ES

PA

NO

L

FR

AN

ÇA

IS

ciency; and if it sells itselfincreasingly as a laboratoryfor developing and testingcreative new ideas on ruraldevelopment, disseminat-ing them, and helping oth-ers to learn its skills, thenboth the continued poverty

challenge within the NAC andthese public good aspects of itswork constitute a strong case forcontinued international donorassistance.

There are also challengeswith sustaining incentives forcollective action in what willinevitably become more pluralis-tic, function-based communityorganizations. With the low fre-quency of repeat traditional infra-structure investments (partly aresult of the AKRSP’s own strat-egy of limiting grants for infra-structure) and increased costrecovery, and with the floating ofthe new microfinance bank, theAKRSP and the AKDN family willneed to listen carefully to evolv-ing community needs. To main-tain sustainability, the AKRSP willalso need to forge improved linkswith the private sector, give prior-ity to under-served women andwomen’s groups, further buildskills needed both locally andnationally, and offer naturalresource management technologyto under-served high-altitudeareas and poorer households.Individual programs also facesustainability challenges. Micro-hydel schemes, in particular,need to make a greater effort atcost recovery. In the marketingand enterprise development pro-gram, the diverse activities havebeen uneven in their sustainabil-ity, suggesting the need to priori-tize better in this area.

E x e c u t i v e S u m m a r y

x i x

EN

GL

ISH

zaciones rurales creadas por

el programa han madurado y

probablemente sobrevivirían

sin necesidad de supervisión

estrecha por parte del AKRSP:

“Sería difícil, pero nos las

podríamos arreglar”, en pala-

bras de un líder comunitario.

La infraestructura construida en el

marco del programa es también sos-

tenible, ya que fue elegida por las

aldeas, está bien mantenida y tiene

una rentabilidad positiva. La sosteni-

bilidad del programa de microfinan-

ciamiento ha sido y continúa siendo

aceptable, a pesar de recientes pro-

blemas operacionales, y el programa

se está transfiriendo ahora a un

nuevo banco controlado por el

AKRSP. No obstante, se observan

algunos riesgos importantes en lo

que respecta a la sostenibilidad.

El riesgo más obvio, compartido

con todos los programas sin dota-

ción y respaldados por donantes, es

que el AKRSP no es financiera-

mente sostenible. Depende del con-

tinuado patrocinio de los donantes.

Los signos de fatiga de los donantes

pueden dar lugar a una reducción

del financiamiento en el futuro

(aunque los acontecimientos recien-

tes en la región podrían cambiar la

situación). Si el AKRSP adopta

medidas de austeridad, si persigue

ordenadamente el objetivo de una

coalición para el desarrollo, si ins-

taura una nueva estrategia que pro-

mueva la eficiencia y si se vende

cada vez más como laboratorio

encargado de formular y ensayar

nuevas ideas creativas sobre el des-

arrollo rural, y de difundirlas y ayu-

dar a otros a adquirir conocimientos

prácticos, tanto el continuado des-

afío de la pobreza dentro de las

ZSC como los aspectos de su labor

relacionados con los bienes públi-

cos constituyen un fuerte argu-

tions villageoises créées par

le programme ont atteint la

maturité et survivraient pro-

bablement sans la supervi-

sion rigoureuse du PARAK

— « cela serait difficile, mais

nous y arriverions », a dit un

leader communautaire.

L’infrastructure construite dans le

cadre du programme est également

durable parce qu’elle a été choisie

au niveau du village, est bien entre-

tenue et assure un retour positif sur

l’investissement. La durabilité du

programme de microfinancement

demeure bonne malgré quelques

échecs opérationnels, et le pro-

gramme est en cours de transfert à

une nouvelle banque contrôlée par

le PARAK. Cependant, la durabilité

comporte des risques particuliers.

Le risque le plus manifeste, mais

qui est commun à tous les program-

mes sans dotation soutenus par des

organismes donateurs, est que le

PARAK n’est pas durable sur le plan

financier. Il dépend du patronage

continu des donateurs. Tout signe

de fatigue de la part des organismes

donateurs peut mener à une réduc-

tion de financement dans l’avenir

(bien que les événements survenus

récemment dans la région puissent

changer la situation). Si le PARAK

« se serre la ceinture », s’il poursuit

vigoureusement le but de former

une coalition de développement,

s’il développe une nouvelle straté-

gie axée sur l’efficacité, s’il se

fait valoir comme laboratoire d’éla-

boration et d’essai d’idées novatri-

ces en développement rural, s’il

dissémine ces idées et aide les

autres à assimiler ses compétences,

le défi persistant de la pauvreté

dans les ZNC et les aspects de ses

travaux axés sur le bien public for-

ment un argument irréfutable en

faveur de la continuation de l’aide

ES

PA

NO

L

FR

AN

ÇA

IS

Lessons for theDevelopmentCommunityThe harsh and oftenremote high mountain val-leys of the WesternHimalayas, Karakorams,and Hindukush are among

the most demanding settings inthe world for social and economicdevelopment. About 900,000 peo-ple in most of the 1,100 villagesscattered over the rugged territoryof the NAC are led by commu-nity-based organizations that havebeen inspired and assisted for 18years by the AKRSP. Two thou-sand new irrigation, road, andother construction schemes havebeen completed. Thousands havebeen trained in productive skills,villagers have come together tomanage their own affairs, thou-sands of small loans have beenmade and repaid, and new tech-nology has spread widely.Incomes have risen, welfareimproved, lives made a little eas-ier, and a start made on helpingwomen to realize their potential.As a result, word has spread, andthe highly successful techniquesof the AKRSP are being success-fully used and adapted in similarsocial situations throughout Pak-istan and elsewhere.

The experiences of the AKRSPwill likely continue to influencecommunity development through-out Pakistan, as well as in areasboth nearby (especially in neigh-boring Afghanistan) and fartherafield. This study identified sev-eral lessons worthy of note for theglobal development community.Some are old, none entirely new,but all are worthy of repetition.• In rural areas, broad and sus-

tained gains in economic and

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

x x

EN

GL

ISH

mento para que continúe la

asistencia de los donantes

internacionales.

Hay también problemas

para mantener los incentivos

a una acción colectiva en lo

que llegarán a ser unas orga-

nizaciones comunitarias

basadas en la función y cada vez

más pluralistas. Debido a la baja fre-

cuencia de las inversiones repetidas

en infraestructura tradicional (en

parte, por la propia estrategia del

AKRSP de limitar las donaciones

para infraestructura) y a la mayor

recuperación de costos, así como a

la puesta en marcha del nuevo

banco de microfinanciamiento, el

AKRSP y la AKDN deberán estar

muy atentos a las nuevas necesida-

des de la comunidad. Para mantener

la sostenibilidad, el AKRSP deberá

forjar también vínculos más estre-

chos con el sector privado, dar prio-

ridad a las mujeres desatendidas y a

los grupos de mujeres, desarrollar

los conocimientos prácticos necesa-

rios tanto a escala local como nacio-

nal y ofrecer tecnologías de gestión

de los recursos naturales para las

zonas altas desatendidas y los hoga-

res más pobres. Los programas con-

cretos deben afrontar también

problemas de sostenibilidad. Los

planes de energía hidroeléctrica en

pequeña escala, en particular, deben

realizar un mayor esfuerzo de recu-

peración de costos. En el programa

de comercialización y desarrollo de

empresas, las distintas actividades

han logrado niveles muy diferentes

de sostenibilidad, lo que demuestra

la necesidad de establecer mejor las

prioridades en esta esfera.

Enseñanzas para lasinstituciones de desarrollo Los inhóspitos y remotos valles eleva-

dos del Himalaya occidental, Karako-

des organismes donateurs

internationaux.

Il existe également des

défis au niveau du maintien

des incitations à l’action col-

lective dans les organisations

communautaires qui seront

inévitablement de plus en

plus pluralistes et basées sur les

fonctions. Étant donné la faible fré-

quence des investissements répétés

dans l’infrastructure (due en partie

à la propre stratégie du PARAK

visant à limiter les subventions pour

l’infrastructure), la faible améliora-

tion du recouvrement des coûts et

le flottement de la nouvelle banque

de microfinancement, le PARAK et

le groupe RDAK devront prêter une

attention particulière à l’évolution

des besoins communautaires. Pour

maintenir la durabilité, le PARAK

devra également forger des liens

avec le secteur privé, accorder la

priorité aux femmes mal desservies

et aux groupements féminins, ren-

forcer les compétences nécessaires

aux plans local et national, et offrir

une technologie de gestion des res-

sources naturelles aux régions à

haute altitude sous-desservies et

aux ménages démunis. Les pro-

grammes individuels aussi font face

à des défis de durablité. Les thèmes

Microhydel en particulier doivent

faire un plus grand effort pour amé-

liorer le recouvrement des coûts.

Dans le programme de commercia-

lisation et de développement des

entreprises, la durabilité des diver-

ses activités a été inégale, ce qui

suggère la nécessité d’améliorer

l’établissement des priorités dans ce

domaine.

Leçons pour la communautéde développement Les vallées ingrates et souvent éloi-

gnées dans les chaînes montagneu-

ES

PA

NO

L

FR

AN

ÇA

IS

social welfare must bebased on stable in-creases in output. In-creased output requiresthat appropriate technol-ogy be available tobring about relativelyquick returns.

• There are ranges of intensityand coverage options in ruralprograms from low input/lowoutput to high input/high out-put. (The tradeoffs warrantmore exploration by the devel-opment community.) Theobject should be to maximizereturns to the scarcest resource.Thus, where community facili-tation and program manage-ment skills are scarce, as in theNAC, a high input/high outputstrategy, exploiting programcomplementarities while mak-ing efficient use of these skills,is a sound option.

• Direct, intensive targeting ofthe poorest of the poor, whileuseful for establishingprocesses and understandingneeds, warrants careful moni-toring if it is to be efficient. Itcan be very staff-intensive andmay prove in the end to beless efficient than broaderapproaches.

• Expansion (scaling up) is aslow process even with skilledfacilitators, as there are feweconomies of scale. (The num-ber of households supportedby this well-managed, multi-component program grew atabout 10 percent per year—now, after 18 years, reachingclose to 100,000 households.)However, expansion by graft-ing new programs into loca-tions with similar circumstancesoffers substantial leverage.

ram e Hindukush figuran

entre los lugares más inade-

cuados del mundo para el

desarrollo económico y social.

Unas 900.000 personas de la

mayoría de las 1.100 aldeas

dispersas sobre el accidentado

territorio de las ZSC están

dirigidas por organizaciones comuni-

tarias, inspiradas y orientadas desde

hace 18 años por el AKRSP. Se han

terminado 2.000 nuevos planes de

riego, carreteras y otras obras de

construcción. Miles de personas han

recibido capacitación en técnicas pro-

ductivas, los habitantes han podido

reunirse para resolver sus propios

asuntos, se han otorgado y reembol-

sado miles de pequeños préstamos y

las nuevas tecnologías se han difun-

dido ampliamente. Los ingresos han

aumentado, el bienestar ha mejorado,

las condiciones de vida son un poco

más fáciles y se ha comenzado a ayu-

dar a la mujer a hacer realidad su

potencial. En consecuencia, se ha

divulgado la voz, y las rentables téc-

nicas del AKRSP se están utilizando y

adaptando con provecho en situacio-

nes sociales semejantes de todo

Pakistán y en otros países.

Las experiencias del AKRSP con-

tinuarán influyendo probablemente

en el desarrollo comunitario de

todo el país, así como en regiones

próximas (sobre todo en el país

vecino Afganistán) y en territorios

muy distantes. En el estudio llevado

a cabo se subrayan varias enseñan-

zas dignas de interés para las insti-

tuciones internacionales de

desarrollo internacional. Algunas

son antiguas, ninguna es del todo

nueva, pero en cualquier caso vale

la pena repetirlas.

• En las zonas rurales, los progre-

sos amplios y sostenidos del

bienestar económico y social

deben estar basados en aumen-

ses de l’Himalaya occidental,

du Karakoram et de l’Hindu-

kush sont considérées

comme des milieux parmi

les plus exigeants du monde

en termes de développement

économique et social. Envi-

ron 900 000 personnes qui

habitent dans les 1 100 villages par-

semés sur le territoire accidenté des

ZNC sont gouvernées par des orga-

nisations communautaires inspirées

et aidées pendant 18 ans par le

PARAK. Deux mille nouveaux

ouvrages d’irrigation, des routes et

autres constructions sont terminés.

Des milliers de personnes ont béné-

ficié d’une formation spécialisée, les

villageois se sont réunis pour gérer

leurs propres affaires, des milliers

de petits prêts ont été accordés et

remboursés, et la nouvelle techno-

logie s’est répandue partout. Les

revenus ont augmenté, le bien-être

s’est amélioré, la vie est un peu

plus facile et on commence à aider

les femmes à exploiter leur poten-

tiel. Ainsi, l’effet s’est répandu et les

techniques très réussies du PARAK

sont utilisées avec succès et adap-

tées selon les circonstances sociales

au Pakistan et ailleurs.

Il est vraisemblable que les

expériences du PARAK continueront

d’influencer le développement com-

munautaire dans tout le Pakistan,

ainsi que dans des régions aussi

bien proches (notamment chez son

voisin l’Afghanistan) qu’éloignées.

Cette étude a identifié plusieurs

leçons méritoires pour la commu-

nauté de développement mondiale.

Certaines sont déjà connues,

aucune n’est entièrement nouvelle,

mais toutes valent la peine d’être

dupliquées.

• Dans les zones rurales, les gains

durables doivent être basés sur

un accroissement stable du ren-

E x e c u t i v e S u m m a r y

x x i

EN

GL

ISH

FR

AN

ÇA

IS

ES

PA

NO

L

• Incentives for commu-nity action can changequite rapidly as pro-grams evolve. Programsshould continually ana-lyze and adjust incen-tives, both economicand social, for different

categories of households. • Partnerships—between gov-

ernment, civil society, NGOs,and the private sector—withactionable and measurablecommitments should be for-mally agreed at the start of aprogram to ensure clarity ofroles and to create incentivesfor performance.

• Eventual exit should be aphased process of increasinglydifferentiated but diminishingcommunity support tailored to,and partly by, each commu-nity, with whom such strategiesshould be agreed up front.

Directions for the FutureWhat sort of organization shouldthe AKRSP seek to be in 10 to 15years? Only the AKRSP itself candecide, but the evidence suggestsit should be smaller and moreembedded in an increasingly inte-grated AKDN family. It shouldbecome less indispensable to theNAC—or indispensable in anentirely different way—through asteady shift toward greater govern-ment, corporate sector, and civilsociety partnerships focused onachieving greatly enhanced devel-opment effectiveness among allthe key institutions of the NAC.The AKRSP should become a rela-tively modest-sized player, but in arelatively larger and more institu-tionally pluralistic NAC team. Itshould be as much a think tank fornew development ideas for the

tos estables de la produc-

ción. Para aumentar la

producción se requieren

tecnologías adecuadas

que permitan conseguir

rendimientos relativa-

mente rápidos.

• En los programas rurales,

hay distintas opciones de intensi-

dad y cobertura, desde un bajo

volumen de producción con

escasos insumos hasta una pro-

ducción elevada con abundantes

insumos (las soluciones de com-

promiso entre las distintas posi-

bilidades deben ser objeto de

examen más detenido por las

instituciones de desarrollo). El

objetivo debe ser multiplicar los

rendimientos con el mínimo de

recursos. Por ello, cuando las

técnicas de gestión de programas

y de facilitación comunitaria son

escasas, como ocurre en las ZSC,

puede ser acertada la estrategia

de abundantes insumos/produc-

ción elevada, que aprovecharía

las complementariedades de los

programas al mismo tiempo que

utilizaría de manera eficiente

esas capacidades.

• La concentración directa e inten-

siva en los más pobres, si bien

útil para establecer procesos y

comprender las necesidades,

debe ser objeto de atenta super-

visión para que resulte eficiente.

Puede requerir personal muy

numeroso y, a la larga, ser

menos eficiente que otros plan-

teamientos más amplios.

• La expansión (reproducción en

escala superior) es un proceso

lento aun cuando se disponga de

personal experimentado, ya que

hay pocas economías de escala

(el número de hogares que reci-

ben ayuda de este programa con

numerosos componentes y bien

dement. Le rendement

accru exige une disponibi-

lité de technologies appro-

priées pour assurer un

retour sur l’investissement

relativement immédiat.

• Il existe une grande

gamme d’options d’inten-

sité et de couverture pour les

programmes ruraux, allant de

fables intrants/faibles rende-

ments à hauts intrants/hauts ren-

dements. (L’arbitrage mérite une

exploration plus approfondie par

la communauté de développe-

ment.) L’objet doit être de maxi-

miser les rendements de la

ressource la plus rare. Ainsi, là

où la facilitation communautaire

et les compétences de gestion

des programmes sont rares,

comme c’est le cas dans les ZNC,

une stratégie basée sur le prin-

cipe de hauts intrants/hauts ren-

dements et l’exploitation des

complémentarités des program-

mes tout en assurant l’utilisation

efficace de ces aptitudes, est une

option saine.

• Bien que le ciblage soit utile

pour établir des processus et

comprendre les besoins, le

ciblage direct et intensif des plus

pauvres des pauvres requiert un

suivi minutieux si l’on veut qu’il

soit efficace. Il peut demander

beaucoup de personnel et peut

être, en définitive, moins efficace

que des approches moins fines.

• L’expansion (augmentation gra-

duelle) est un processus lent

même quand on dispose de faci-

litateurs qualifiés, car il existe

peu de possibilité d’économies

d’échelle. (Le nombre de ména-

ges couverts par ce programme à

multiples composantes bien

gérées a augmenté d’environ 10

pour cent par an — maintenant,

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

x x i i

EN

GL

ISH

FR

AN

ÇA

IS

ES

PA

NO

L

NAC and Pakistan as it is anarea program, developingcreative community devel-opment and policy optionsfor government. It shouldaim for intellectual leader-ship in key areas of pro-poor development—for

example, through creative experi-mentation, in a culturally sensitiveway, to bring women more intothe mainstream. It should see eval-uation, learning, and transfer oflessons both within and outsidePakistan as an explicit part of itscapacity development strategy. Itshould recover more of its costs,sell more of its services, and per-haps be supported by a coreendowment. With its AKDN part-ners in particular it should focuseven more on human capacitydevelopment. For its clients, indue course, it should become ademand-led service organization,ultimately relying on vastlyimproved telecommunications andbetter main roads. While currentevents may disrupt for a while, theAKRSP must paint on a broadercanvas—a canvas that is informedby, but stretches well beyond, itspast vision of creating and sustain-ing village organizations.

RecommendationsThe AKRSP’s comparative advan-tages lie mainly in communityfacilitation, program manage-ment, and human resource devel-opment at the village level; thelinking of professional skills withlocal community skills; analyticalskills related to monitoring andevaluation; to some extent, pol-icy; and, above all, the capacityto “pull it all together.” Thesecomparative advantages can beexploited by a strategy that

administrado creció apro-

ximadamente un 10% al

año: ahora, después de 18

años, se habría llegado a

un total próximo a los

100.000 hogares). No obs-

tante, la expansión basada

en la introducción de

nuevos programas en lugares

con circunstancias semejantes

puede tener un considerable

efecto multiplicador.

• Los incentivos para la acción

comunitaria pueden cambiar

rápidamente, a medida que evo-

lucionan los programas. Éstos

deben analizar y acomodar cons-

tantemente los incentivos, tanto

económicos como sociales,

teniendo en cuenta las diferentes

categorías de hogares.

• Las asociaciones —entre el gob-

ierno, la sociedad civil, las ONG y

el sector privado— con compro-

misos aplicables y cuantificables

deberían aprobarse por consenso

al comienzo del programa, para

garantizar la claridad de funciones

e incentivar el desempeño.

• La estrategia de salida debería

ser un proceso gradual de apoyo

comunitario cada vez más dife-

renciado pero decreciente, de

acuerdo con las necesidades de

cada comunidad, con la que

deberían aprobarse inicialmente

dichas medidas.

Orientaciones para el futuro¿Qué tipo de organización debería

tratar de ser el AKRSP dentro de 10

a 15 años? Únicamente el AKRSP

puede decidirlo, pero los testimo-

nios disponibles parecen indicar que

debe reducir su tamaño e incorpo-

rarse progresivamente a una familia

AKDN cada vez más integrada.

Debería ser menos indispensable

para las ZSC —o indispensable de

après 18 ans, il couvre

près de 100 000 ména-

ges.) Cependant, l’expan-

sion par greffe de

nouveaux programmes

dans des contextes simi-

laires présente de grands

avantages.

• Les incitations à l’action commu-

nautaire peuvent changer rapide-

ment à mesure que les

programmes évoluent. Les pro-

grammes doivent analyser et

adapter continuellement les inci-

tations d’ordre social et économi-

que pour les diverses catégories

de ménages.

• Les partenariats — entre le gou-

vernement, la société civile, les

ONG et le secteur privé —

accompagnés d’engagement

donnant droit d’action et de

mesure devraient être conclus

formellement dès le début d’un

programme afin d’assurer la défi-

nition claire des rôles et de créer

des incitations à la performance.

• Le désengagement éventuel

échelonné de l’appui commu-

nautaire devrait être un proces-

sus de plus en plus différencié

mais déclinant, adapté à, et en

partie par, chaque communauté

avec laquelle ces stratégies

avaient été convenues au départ.

Lignes Directrices FuturesQuelle sorte d’organisation le PARAK

doit chercher à devenir d’ici 10 à 15

ans ? Seul le PARAK peut en décider,

mais l’évidence laisse à penser qu’elle

devrait être de moindre envergure et

mieux imbriquée dans le groupe

RDAK de plus en plus intégré. Elle

devrait devenir moins indispensable

dans les ZNC — ou indispensable

d’une toute autre façon — grâce à une

réorientation soutenue vers de plus

grands partenariats avec le gouverne-

E x e c u t i v e S u m m a r y

x x i i i

EN

GL

ISH

FR

AN

ÇA

IS

ES

PA

NO

L

addresses the followingrecommendations.

Fine-tune the approachto improving rural NAClivelihoods.

• Ensure attention to the poor and women in the neworganizations. The shift to amore pluralistic, functions-based approach to communityorganizations is an appropriateand inevitable evolution, but itcarries some risks for vulnera-ble people.

• Remain fully engaged in micro-finance. The new microfinancebank notwithstanding, theAKRSP should ensure: (i) acontinued explicit microfinancestrategy, covering the povertyobjective in particular; (ii) fieldcoordination with the bank andmaintenance of the importantlinkages between microfinanceand other AKRSP programactivities; and (iii) clear agree-ment on how the very riskyemerging internal lending willbe supervised by the new bankwith AKRSP support.

• Improve efficiency throughdifferentiation of supportaccording to individual com-munity needs—more for poor,more vulnerable communities,less for mature communities.This would call for villageconsultations, the develop-ment of a classification system,and guidelines for staff ongraduation criteria andprocedures.

• Increase cost recovery, reviewthe grant/subsidy strategy, andprepare criteria and guidelinesto ensure optimization ofgrant/subsidy impact.

una forma totalmente

nueva— gracias a un cons-

tante desplazamiento hacia

una mayor asociación entre

el gobierno, las empresas y la

sociedad civil, con el objetivo

de lograr una mayor eficacia

en términos de desarrollo

entre todas las instituciones clave de

las ZSC. El AKRSP debería conver-

tirse en un protagonista relativa-

mente modesto, pero dentro de un

equipo de las ZSC relativamente

mayor e institucionalmente más plu-

ralista. Debería ser tanto un grupo

de reflexión sobre los nuevos con-

ceptos de desarrollo para las ZSC y

Pakistán como un programa geográ-

fico, que permita formular opciones

creativas sobre políticas de desarro-

llo comunitario para el gobierno.

Debería tratar de desempeñar una

función de liderazgo intelectual en

esferas clave del desarrollo favorable

a los pobres —por ejemplo, medi-

ante la experimentación creativa, en

forma culturalmente flexible, para

integrar más a la mujer. Debería

considerar la evaluación, el aprendi-

zaje y la transferencia de enseñanzas

tanto dentro como fuera de Pakistán

como parte expresa de su estrategia

de desarrollo de la capacidad. Debe-

ría recuperar mayor parte de sus

costos, vender más servicios y,

quizá, contar con una dotación

básica. En particular, junto con sus

socios de la AKDN, debería cen-

trarse todavía más en el desarrollo

de la capacidad humana. Para sus

clientes, cuando llegue el momento

oportuno, debería convertirse en

una organización de servicios impul-

sados por la demanda, basada en

último término en servicios de tele-

comunicaciones inmensamente

mejorados y en una red de carrete-

ras principales de mayor calidad. Si

bien los actuales acontecimientos

ment, le secteur commercial

et la société civile, partena-

riats axés sur l’amélioration de

l’efficacité des institutions clés

des ZNC. Le PARAK devrait

devenir un acteur relativement

modeste, mais faire partie

d’une équipe ZNC plus grande

et davantage pluraliste au plan insti-

tutionnel. Il devrait jouer aussi bien le

rôle de groupe de réflexion chargé de

trouver de nouvelles idées de déve-

loppement pour les ZNC et le Pakis-

tan que le rôle de programme régional

chargé du développement commu-

nautaire novateur et de l’élaboration

d’options politiques pour le gouver-

nement. Il devrait faire appel au lea-

dership intellectuel pour les zones

clés de développement en faveur des

pauvres — par exemple à travers

l’expérimentation novatrice en ten-

ant compte de la sensibilité culturelle

dans le but d’intégrer davantage les

femmes dans le processus général. Il

devrait considérer l’évaluation,

l’apprentissage et le transfert des

leçons aussi bien à l’intérieur qu’à

l’extérieur du pays comme partie

explicite de sa stratégie de dévelop-

pement des capacités. Il devrait recou-

vrer une plus grande partie de ses

coûts, offrir une plus grande quantité

de services et peut-être bénéficier

d’une dotation de base. Avec ses par-

tenaires du groupe RDAK en particu-

lier, il devrait centrer ses efforts

davantage sur le développement des

capacités. Pour ses clients, il devrait

devenir, en temps utile, une organi-

sation de services régie par la

demande, qui en bout de ligne, pour-

rait s’appuyer sur des services de télé-

communication nettement améliorés

et un réseau de meilleures routes

principales. Bien que les événements

actuels puissent demeurer perturbants

pendant quelque temps encore, le

PARAK doit peindre sa vision sur une

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

x x i v

EN

GL

ISH

FR

AN

ÇA

IS

ES

PA

NO

L

• Increase the gender con-tent in data collection inboth future socio-economic surveys andmicrofinance monitoringto improve understand-ing of gender issues.

• Carefully monitor thepoverty pilot to test differentlevels of staff and resourceintensity so that relative povertyalleviation efficiency in compari-son with the core program canbe evaluated.

Increase institutional capacityto sustain development forthe long term throughpartnerships.

• Take the lead within the AKDNin developing a coalition withgovernment. This coalitionshould aim to raise significantlygovernment’s developmentcapacity in a specified period—with an increasing role for thecorporate sector and otherNGOs. (A workshop of poten-tial partners leading to workingcommittees would be one wayto start.)

• Agree and implement a fullycoordinated approach to devel-opment in the NAC with theother organizations in theAKDN, including looking atopportunities for cost sharing.Strongly support the incipientshift to devolution throughcapacity building (especiallyfor women), policy work, anda joint monitoring partnershipwith government. Monitoringwould feed back as the processmoves forward. It wouldinclude monitoring and evalua-tion capacity development forgovernment.

pueden representar un tras-

torno provisional, el AKRSP

debe adoptar una perspectiva

más amplia, inspirada en el

concepto anterior de crea-

ción y sostenimiento de orga-

nizaciones rurales, pero en

un contexto mucho más

amplio.

RecomendacionesLas ventajas comparativas del

AKRSP consisten principalmente en

la facilitación comunitaria, la ges-

tión de programas, el desarrollo de

los recursos humanos en las aldeas,

la conexión entre capacidad profe-

sional y capacidad comunitaria, los

conocimientos analíticos relaciona-

dos con el seguimiento y la evalua-

ción, las políticas —en cierto

sentido— y, por encima de todo, la

capacidad de “englobarlo todo”.

Estas ventajas comparativas pueden

aprovecharse mediante una estrate-

gia que tenga en cuenta las siguien-

tes recomendaciones.

Ajustar el planteamiento paramejorar los medios de vida de lasZSC rurales.• Garantizar la atención a los

pobres y las mujeres en las nue-

vas organizaciones. La orientación

hacia un concepto de organiza-

ción comunitaria más pluralista y

basado en las funciones es una

evolución adecuada e inevitable,

pero conlleva ciertos riesgos para

los grupos vulnerables.

• Mantener plenamente el compro-

miso con el microfinanciamiento.

A pesar de contar con un nuevo

banco de microfinanciamiento,

el AKRSP debería ofrecer i) una

estrategia continuada y explícita

de microfinanciamiento, que

abarque en particular el objetivo

de la pobreza; ii) coordinación

plus grande toile de fond —

une toile de fond basée sur sa

première vision, mais tendue

au-delà de la création d’orga-

nisations villageoises durables.

RecommandationsLes avantages comparatifs du

PARAK portent surtout sur la facili-

tation communautaire, la gestion du

programme, le développement des

ressources humaines au niveau

villageois, le lien entre les compé-

tences professionnelles et les com-

pétences communautaires locales,

les compétences analytiques en

matière de suivi et d’évaluation, la

politique dans une certaine mesure

et, surtout, la capacité d’orchestrer

tous ces éléments. Ces avantages

comparatifs peuvent être exploités

par le biais d’une stratégie qui

aborde les recommandations

suivantes :

Affiner l’approche en amélio-rant les moyens de subsistancerurale dans les ZNC.

• Prêter attention aux pauvres et

aux femmes dans les nouvelles

organisations. La réorientation

vers une approche davantage

pluraliste et basée sur les fonc-

tions des organisations commu-

nautaires représente une

évolution appropriée et inévita-

ble, mais elle comporte des ris-

ques au niveau des populations

vulnérables.

• Rester entièrement engagé dans

le microfiancement. Nonobstant

la nouvelle banque de microfi-

nancement, le PARAK devrait

assurer la mise en place : a)

d’une stratégie de microfinance-

ment explicite visant en particu-

lier à atteindre l’objectif de

pauvreté ; b) des activités de

coordination sur le terrain avec

E x e c u t i v e S u m m a r y

x x v

EN

GL

ISH

FR

AN

ÇA

IS

ES

PA

NO

L

Maximize the leverage ofthe acquired AKRSPskills both within Pak-istan and externally.

• Develop an explicitstrategy to guide theAKRSP’s approach to

sharing its knowledge bothinside and outside Pakistan, setgoals, and monitor achieve-ments as with any other pro-gram component.

• Become leaders in Pakistan onthe gender issue. Look at com-parisons with other programsin the area of gender. Appointa leader for the women’s pro-gram at a senior level andexperiment with a gradual shiftaway from the separatewomen’s and men’s organiza-tions, and toward the inclusionof women in overall commu-nity decisionmaking. The DehiCouncils, depending on howthey evolve, may present anopportunity for this throughtheir women’s membership.

Donors need to standardizemonitoring and evaluation for-mats and processes and acceptmulti-donor evaluation reports.

A large amount of highly skilledAKRSP staff time is taken up try-ing to meet different requirementssuch as donor-specific log-frameformats. These resources couldbe better spent pursuing impor-tant evaluative questions such asthe optimal strategy for cost-effectiveness in poverty allevia-tion or monitoring and givingfeedback to government on theevolution of Dehi Councils.

sobre el terreno con el

banco y mantenimiento

de vínculos importantes

entre el microfinancia-

miento y otras actividades

del programa AKRSP, y

iii) un claro acuerdo

sobre la forma en que el

nuevo banco supervisará, con

apoyo del AKRSP, los nuevos

préstamos internos, que presen-

tan un alto nivel de riesgo.

• Mejorar la eficiencia mediante la

diferenciación del apoyo de

acuerdo con las necesidades

comunitarias individuales —más

en favor de los pobres, más

para las comunidades vulnera-

bles y menos para las comuni-

dades maduras. Para ello habría

que entablar consultas en las

aldeas, elaborar un sistema de

clasificación y ofrecer al perso-

nal orientaciones sobre los crite-

rios y procedimientos de

graduación.

• Aumentar la recuperación de

costos, examinar la estrategia de

donaciones/subsidios y preparar

criterios y orientaciones para

garantizar el máximo efecto de

las donaciones y subsidios.

• Aumentar la atención a las dife-

rencias entre los sexos en la

recopilación de datos tanto en

las futuras encuestas socioeconó-

micas como en la supervisión

del microfinanciamiento, con el

fin de llegar a una mejor com-

prensión de esas cuestiones.

• Supervisar atentamente la expe-

riencia piloto sobre la lucha con-

tra la pobreza con el fin de

comprobar los diferentes niveles

de dotación de personal y los

recursos para poder evaluar la

eficiencia relativa en el alivio de

la pobreza, en comparación con

el programa básico.

la banque et le maintien

des liens importants entre

le microfinancement et

les autres activités du

PARAK ; et c) d’un accord

bien articulé sur la super-

vision des nouveaux prêts

internes à haut risque par

la nouvelle banque avec le con-

cours du PARAK.

• Améliorer l’efficacité à travers la

différentiation de l’appui selon

les besoins communautaires indi-

viduels — plus d’appui aux pau-

vres communautés vulnérables,

moins d’appui aux communautés

matures. Ceci demandera des

concertations au niveau du vil-

lage, l’élaboration d’un système

de classification et des directives

à l’usage du personnel sur les

critères de gradation et sur les

procédures.

• Accroître le taux de recouvre-

ment des coûts, examiner la

stratégie relative aux dons/sub-

ventions, et préparer des critères

et des directives pour assurer

l’optimisation de l’impact des

dons/subventions.

• Pour la collecte des données,

tenir mieux compte de la sexos-

pécificité tant lors des enquêtes

socioéconomiques que du suivi

du microfinancement, afin de

mieux comprendre la probléma-

tique hommes-femmes.

• Surveiller minutieusement le pro-

jet pilote sur la pauvreté afin de

tester les divers niveaux de per-

sonnel et l’intensité des ressour-

ces et d’évaluer l’efficacité de la

réduction de la pauvreté par rap-

port au programme de base.

Augmenter la capacité institu-tionnelle à maintenir le dévelop-pement dans le long terme àtravers les partenariats.

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

x x v i

EN

GL

ISH

FR

AN

ÇA

IS

ES

PA

NO

L

Aumentar la capacidad ins-titucional para respaldar eldesarrollo a más largoplazo mediante relacionesde asociación.• Tomar la iniciativa dentro

de la AKDN para estable-

cer una coalición con el

gobierno. Esa coalición debería

tratar de aumentar significativa-

mente la capacidad gubernamen-

tal de desarrollo en un período

previamente especificado —con

una intervención creciente de las

empresas y otras ONG (una

manera de comenzar sería orga-

nizar un taller de posibles aso-

ciados, que podría dar lugar a

comités de trabajo).

• Aprobar y aplicar un concepto

de desarrollo plenamente coordi-

nado en las ZSC, junto con otras

organizaciones de la AKDN, lo

que supondría también conside-

rar las oportunidades de distribu-

ción de costos. Debería apoyarse

decididamente el nuevo despla-

zamiento hacia la delegación de

responsabilidades mediante el

fortalecimiento de la capacidad

(sobre todo para la mujer), los

estudios sobre políticas y una

asociación con el gobierno para

las actividades de seguimiento.

Los resultados de este segui-

miento deberían aplicarse a

medida que avance el proceso.

En este contexto debería

incluirse el desarrollo de la capa-

cidad de seguimiento y evalua-

ción por parte del gobierno.

Incrementar el efecto multiplica-dor de las técnicas adquiridaspor el AKRSP tanto dentro dePakistán como en el exterior.

• Formular una estrategia explícita

para orientar al AKRSP a com-

partir sus conocimientos tanto

• Prendre la tête de file du

groupe RDAK pour établir

une coalition avec le gou-

vernement. Cette coalition

devrait tendre à rehausser

sensiblement la capacité

gouvernementale pendant

une période déterminée

— le secteur des entreprises et

d’autres ONG joueraient un plus

grand rôle. (On pourrait com-

mencer par organiser un atelier

de partenaires potentiels qui

conduirait à l’établissement de

comités de travail.)

• Convenir d’une approche du

développement dans les ZNC et

l’exécuter de manière bien coor-

donnée, y compris la recherche

d’opportunités de partage des

coûts. Appuyer à ses débuts la

réorientation vers le désengage-

ment à travers le renforcement

des capacités (des femmes en par-

ticulier), le travail d’élaboration

des politiques et un partenariat de

suivi conjoint avec le gouverne-

ment. Le suivi permettrait de ren-

dre compte du déroulement du

processus. Il comprendrait le

développement de la capacité

gouvernementale en matière de

suivi et d’évaluation.

Maximiser le développement descompétences du PARAK acquisesaussi bien au Pakistan qu’à l’extérieur.

• Élaborer une stratégie explicite

guidant l’approche du PARAK en

ce qui concerne le partage de

ses connaissances aussi bien au

Pakistan qu’en dehors, établir les

objectifs et suivre les réalisations

comme pour tout autre compo-

sante d’un programme.

• Devenir les leaders de la problé-

matique hommes-femmes au

Pakistan. Nommer un leader de

E x e c u t i v e S u m m a r y

x x v i i

FR

AN

ÇA

IS

ES

PA

NO

L

dentro como fuera de

Pakistán, fijar objetivos y

supervisar los logros,

como cualquier otro com-

ponente del programa.

• Convertirse en pioneros

de la defensa de la mujer

en Pakistán. Buscar com-

paraciones con otros programas

en lo que se refiere a la relación

entre el hombre y la mujer; nom-

brar un dirigente de rango supe-

rior para el programa en favor

de la mujer y experimentar con

un abandono gradual de las

organizaciones separadas para

hombres y mujeres, en favor de

la inclusión de estas últimas en

los procesos generales de toma

de decisiones de la comunidad.

Los consejos de Dehi, según

cómo evolucionen, pueden ofre-

cer una oportunidad para ello a

través de la participación de las

mujeres como miembros.

Los donantes deben uniformarlos formatos y procesos deseguimiento y evaluación y acep-tar informes de evaluación devarios donantes.

• Un elevado número de emplea-

dos especializados del AKRSP

deben dedicar su tiempo a resol-

ver diferentes trámites, como los

formatos del marco lógico espe-

cíficos de cada donante. Estos

recursos podrían emplearse más

provechosamente tratando de

resolver cuestiones importantes

de evaluación, como la estrategia

óptima para una mayor eficacia

en función de los costos en la

reducción de la pobreza o la

supervisión e intercambio de

opiniones con el gobierno sobre

la evolución de los consejos de

Dehi.

niveau supérieur pour le

programme des femmes

et tester une réorientation

graduelle des organisa-

tions séparées de femmes

et d’hommes vers l’inclu-

sion des femmes dans le

processus général de

prise de décisions communautai-

res. Selon la façon dont ils évo-

luent, les Conseils Dehi

pourraient présenter une oppor-

tunité à ce sujet à travers leurs

membres féminins.

Recommander aux donateurs denormaliser les formats et pro-cessus de suivi et d’évaluation, etaccepter les rapports d’évalua-tion des multi-donateurs. Le personnel hautement qualifié du

PARAK passe une grande partie de

son temps à essayer de satisfaire les

diverses exigences, comme par

exemple les formats d’enregistre-

ment à cadre des organismes dona-

teurs particuliers. Ces ressources

pourraient être plus judicieusement

utilisées en essayant de répondre à

des questions évaluatives importan-

tes telles que la stratégie optimale

ou le suivi de la réduction de la

pauvreté efficace par rapport au

coût, ou encore le compte rendu au

gouvernement de l’évolution des

Conseils Dehi.

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

x x v i i i

FR

AN

ÇA

IS

ES

PA

NO

L

x x i x

AKCS Aga Khan Cultural ServiceAKDN Aga Khan Development NetworkAKES Aga Khan Education ServiceAKHS Aga Khan Health ServiceAKF Aga Khan FoundationAKRSP Aga Khan Rural Support ProgramAKRSP IV Aga Khan Rural Support Program, Phase IVBACIP Building and Construction Improvement ProgramCAS Country Assistance StrategyCIMMYT Centro Internacional para Mejoramiento de Maiz y Trigo (International Maize and

Wheat Improvement Center)DFID Department for International Development (U.K.)ECP Enterprise Credit ProgramERR Economic rate of returnFAO Food and Agriculure OrganizationFMU Field Management UnitGBTI Ghaza Barotha Taragiati IdaraICIMOD International Center for Integrated Mountain DevelopmentIDG International Development GoalIFAD International Fund for Agricultural DevelopmentIUCN International Union for the Conservation of NatureKARINA Karakoram Agricultural Research Institute for the Northern AreasLDO Local development associationLSU Learning Support UnitMDGs Millennium Development GoalsM&E Monitoring and evaluationMER Monitoring, Evaluation, and Research SectionMIES Mountain Infrastructure Engineering ServicesMIS Management information systemMLURI Macaulay Land Use Research InstituteNAC Northern Areas and ChitralNARC National Agricultural Research CenterNGO Nongovernmental organizationNORAD Norwegian Agency for Development CooperationNRM Natural resource managementNRSP National Rural Support ProgramOED Operations Evaluation Department, World BankPAF Poverty Alleviation FundPPB Participatory plant breedingPPI Productive physical infrastructure PRSP Punjab Rural Support ProgramPVS Participatory variety selectionR&D Research and developmentRPM Regional program managerRPO Regional Program OfficeSAM Social accounting matrix

ABBREVIATIONS AND ACRONYMS

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

x x x

SAPAP South Asia Poverty Alleviation ProgramSDI Subsidy Dependence IndexSMS Safe minimum standardsSRSC Sarhad Rural Support CorporationUNDP United Nations Development ProgramVO Village organizationWASEP Water and Sanitation Education ProgramWO Women’s organizationWSO Women’s social organizerWWF Worldwide Fund for Nature

1

Introduction

The purpose of this evaluation was to assess the development out-come of both the full 18-year Aga Khan Rural Support Program(AKRSP) and the 5-year period since the last evaluation by the

Operations Evaluation Department of the World Bank. The performancecriteria that have been used in the evaluation are relevance, efficacy,efficiency, institutional development impact, and sustainability.1 The eval-uation, requested by the Aga Khan Foundation (AKF) and the AKRSP,focused on four program components: community organizations, infrastructuredevelopment, natural resource management, and microfinance.

Given the timing—just before the next donorfunding cycle—the evaluation team was askedto recommend future directions.

AKRSP Program ObjectivesThe original 1981 program proposal included thefollowing statement: “with a program intendedto affect the development of a region, its timehorizon cannot be limited to two or three years.On the other hand, a primary purpose of theAKRSP should be to work itself out of a job.”2

Two years later, the first (1983) strategy paperarticulated the program objectives as “to de-velop an innovative replicable model by a small[nongovernmental organization] acting as a cat-alyst of rural development through workingwith local people to identify and appraise proj-ect opportunities, to promote the provision of

needed services for tackling problems of highmountain areas.” By 1986, the objective had be-come “a doubling of the (rural) per capita incomeover a period of 10 years.” Later objectives state-ments tended to focus on enhancing people’s ca-pacity. Thus, the current mission statement is: “Toenhance the capacity of the peoples of theNorthern Areas and Chitral to sustain and im-prove the quality of their lives. In particular, di-versified, sustainable and equitable economicdevelopment will be promoted through build-ing up the competence and confidence of localorganizations/institutions and individuals, andthrough the provision of economic resources andopportunities.”

It is evident that the trend in the objectivesstatements has been away from both “workingitself out of a job” and physical achievements,

11

and toward capacity building. The earlier focusboth of the AKRSP and the Aga Khan Develop-ment Network (AKDN, see box 1.1) as a wholewas in the Upper Hunza Valley—predominantlyIsmaili communities. However, the intent fromthe outset was to serve all communities.

MethodsThe methods used in this evaluation consistedof the following main elements:• Group and individual discussions in a sample

of 24 villages (see list of villages in Annex F).These were purposively selected from a framedeveloped by the AKRSP at the request of theevaluation team to achieve a representativespread of village types. The elements includedregion, district, Field Management Unit, andcropping zone (single-cropping, double-cropping, transitional) locational data; reli-gious sect; rated relative strength of village

institutions; rated intensity of AKRSP support;and rated relative strength of resource base.The sample included one non-AKRSP villageand one with very little activity.

• The application of four semiformal, partlyquantitative, partly qualitative questionnaireson community organizations, infrastructure,natural resource management, and microfi-nance. The purpose was to ensure consistencyand coverage rather than to produce statisti-cally significant analysis. The mission proce-dure within villages typically involved about90 minutes of discussion at a meeting of thefull village, followed by splitting up for a fur-ther 90 minutes to pursue small group or in-dividual discussion, including discussion withgroups of women, and visits to infrastructureand natural resource management sites.AKRSP staff were present for translation pur-poses most of the time, although the team had

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

2

The AKRSP is a private, noncommunal (that is, supporting all re-ligious sects), nonprofit company established in 1982 by the AgaKhan Foundation. Its operations in the Northern Areas and Chi-tral are funded by a consortium of bilateral and multilateraldonors with the objective of improving the quality of life of thepeople. It supports economic and institutional development oflocal communities in collaboration with government, electedbodies, and other development agencies. The major componentsof the program are social organization, women’s development,natural resource management, development of physical infra-structure, human resource development, enterprise promotion,and credit and savings services. The program reaches some900,000 people in about 1,100 villages and, over the past 5 years,has had an annual budget of about US$6 million. (Annex D givesa timeline of main developments over the program period.)

The process at the village level typically has involved theformation of a village organization (VO) with a members’ sav-ings program, technical help, and grant support for an initialphysical infrastructure project chosen and constructed by theVO (both as an incentive for community action and for its de-velopment impact), followed by support for improving agri-culture, livestock, and forestry productivity and credit servicesfor mainly agricultural inputs. In recent years, second infra-

structure investments have been sponsored, including largerschemes by clusters of VOs. Women’s organizations are es-tablished to serve the special needs of women. Training of vil-lagers has been an important element of the program. Thesavings and credit services are in the process of being devolvedto a new bank.

The Aga Khan Development Network (AKDN) is a world-wide concept spanning all the Aga Khan development activities.It comprises three main groups of activities—economic devel-opment, social development, and culture. In Pakistan, the AKDNis generally taken to mean the AKRSP, Aga Khan Education Ser-vice (AKES); Aga Khan Health Service (AKHS); Aga Khan Build-ing and Construction Improvement Service (BACIP), the largestpart of which is the Water and Sanitation Program (WASEP); AgaKhan Cultural Service (AKCS); and the Aga Khan University inKarachi. The Aga Khan Foundation (AKF) has primary responsi-bility for the AKRSP, including helping to raise funds. AKF alsoassists the other agencies in fund raising. Some of the institu-tions are very old: the AKHS was established in 1923 and the AKESin 1946. The AKRSP commands an annual budget for developmentand operating costs of about Rs. 350 million, while the AKES hasRs. 90 million, the AKHS about Rs. 75 million, and WASEP aboutRs. 50 million. (Exchange rate in 2001 = Rs. 59: US$1.)

A K R S P O p e r a t i o n s a n d T h e i r I n s t i t u t i o n a lS e t t i n g

B o x 1 . 1

two members who spoke Urdu. The impres-sion was of generally very open discussionswith ranges of views expressed.

• Review of the extensive literature, includingthe reports of previous evaluation missionsand the report of the Joint Review Missions(see Annexes B and M), and literature that wascritical as well.

• Discussions with government both at the cen-ter and in the Northern Areas and Chitral(NAC), the Donor Liaison Group, the AKRSPBoard, other institutions, banks, members ofthe AKDN family, and nongovernmental or-ganizations (NGOs) such as the InternationalUnion for the Conservation of Nature (IUCN).

• An economic analysis drawn from case stud-ies and data provided by the AKRSP. This wasbacked up by questions in villages visited onland value changes. The natural resourcemanagement coefficients used were reviewedby two specialists outside the AKRSP andoutside the evaluation team.

• The application of an institutional survey tostaff in the AKRSP at grade 4 and above toassess staff attitudes toward AKRSP goals;perceptions on competence, managementprocesses, and procedures; motivation; re-cruitment practices; and management open-ness. This survey was conducted using sealedenvelopes to protect anonymity.

• A tabulation of cost comparators with otherrural development programs, including dis-cussions with the Pakistan National RuralSupport Program and exchange of informa-tion with the International Fund for Agricul-

tural Development (IFAD) and the IndiaAKRSP.

• The use of industry standard comparators inmicrofinance.

• A modest amount of further analysis of the1991 and 1997 socio-economic survey data,mainly aimed at seeking further evidence onattribution.

• The development of a “timeline” showingthe main changes by program componentsince 1983.

• A workshop with AKRSP microfinance staff,conducted by the evaluation team.

Findings of Previous EvaluationsAnnex B outlines the recommendations ofthe three previous OED evaluations and AnnexM the recommendations of the main recentdonor evaluations. Overall, adoption of those rec-ommendations has been partial. A number ofrecommendations have been repeated severaltimes in some form and feature again in this eval-uation. These include the need to strengthenmarketing linkages, the farm systems recom-mendations in their various forms, the need toaddress productive packages for women, and theneed for improved linkages with other agencies.Reasons for the failure to adopt earlier recom-mendations include concerns about cost-effectiveness, skill gaps (for example, in thegender area), donor funding constraints, thedifficulty of finding productive opportunitiesand accessing markets in such a challengingarea, and limited specificity of the recommen-dations themselves.

I n t r o d u c t i o n

3

5

Overall ProgramAssessment

The purpose of this section is to (i) provide an overall assessment ofthe program focusing on relevance, efficacy, and efficiency; (ii) placethe AKRSP in the context of the institutional performance of the NAC

as a whole; and (iii) evaluate the AKRSP as an organization. The focus ison evaluating achievements.

Recommendations to address the problemsidentified are developed in Chapter 5, “FutureDirections.”

Relevance Overall, the AKRSP’s achievement has beenremarkable. It has remained substantially relevantto the overall goals of development in Pakistan,to the changing needs of the Northern Areas andChitral (NAC), and to the objectives and missionstatements as they have evolved. The AKRSP hassought to reduce poverty and improve peoples’lives through broad-based rural development inan area where income levels are still below thepoverty target set in the Millennium DevelopmentGoals (MDGs).1 (By latest World Bank estimates,the poverty incidence in the rural areas is 37.9 per-cent for the Northern Areas and 46.5 percent forNorth West Frontier Province, placing both nearthe bottom of the Pakistan province/region rank-ing.2) However, there is now some danger ofreduced relevance arising, not so much from fail-ures by the AKRSP as from its successes, combined

with a changing external environment. Success-ful social and economic development have givenrise to growing demands for improved services,and the capacity to articulate those demands isincreasing pressure on both the AKRSP and gov-ernment. Moves within the Aga Khan Develop-ment Network (AKDN) in Pakistan to enhancecoordination and the expansion of mandates ofthose members with less coverage beyond theIsmaili community offer new strategic challenges.Funding prospects have been shrinking, andsome donors have been pressing for reductionsin scale and greater differentiation aimed at oper-ational efficiency. If the AKRSP is to maintain itsrelevance and improve its development per-formance, it must respond successfully to thesechanging circumstances.

The AKRSP’s Comparative AdvantageWithin Pakistan and the NAC, the AKRSP’s com-parative advantage is rooted in its managerialexpertise; its educated, skilled staff, mostly drawnfrom all over the program area; the knowledge

22

and contacts necessary for it to draw on outsideexpertise; its creativity in development and socialorganization; its 20 years of intensive experience;and its ability to mobilize funds. From the peo-ples’ perspective, the AKRSP is the most com-petent development agent in the NAC. The teamrecorded frequent and eloquent testimony to thiseffect at many villages throughout the NAC.This is recognized by government, whichrecently called on the AKRSP to help set up DehiCouncils as a first stage in the reform of localgovernment.3 The numerous village organizationsfostered by the AKRSP (Annex J, table J.4) arenow the foundation for other agencies’ work,including government, the International Unionfor the Conservation of Nature (IUCN), andother members of the AKDN.

Within the AKDN, the AKRSP’s comparativeadvantage lies in its rural focus, the goodwill itcommands within government and into the far-thest corners of the region, and its demonstratedorganizational and managerial talents. More-over, the AKRSP has more experience than mostother members of the AKDN in working withnon-Ismaili communities. It is also many timeslarger, measured by budgetary resources or geo-graphic coverage, than any of the other AKDNmembers in Pakistan. These strengths are rec-ognized within the AKDN. However, recognitiondoes not translate into a ready willingness toallow the AKRSP to lead, much less dominate,the AKDN. The AKES and AKHS (see box 1.1)both have much longer traditions of working inPakistan than the AKRSP does, and throughtheir separate governing boards they have givenrise to well-established ideas and procedures, andnot a little resistance to change.

The AKRSP and GovernmentAlthough the AKRSP has an overwhelming pres-ence in rural parts of the Northern Areas, it cannever become a viable alternative to governmentwith its far broader responsibilities. However, inrural development, the AKRSP has effectivelysubstituted for government.4 The programembraces 98 percent of villages, where it pro-vides many services usually supplied by gov-ernment agencies (see Annex J, table J.1 fordetails of program coverage and achievements).For the most part, ordinary people regard theAKRSP with respect and gratitude as a guide andsupporter, particularly in the Northern Areas,where the population lives under special polit-ical arrangements (box 2.1). In urban areas, insuch aspects of development as large infra-structure, and in policymaking, the AKRSP’sinfluence is modest.5

The evaluation team found no evidence ofgovernment activity to compare with the range,depth, and quality of the AKRSP’s rural activi-ties. Villagers commonly claim that they have hadno contact with government agricultural officialsfor many years. Government officials themselveshave difficulty remembering when they lastengaged in fieldwork, having little means orincentive to leave their stations. Department ofAgriculture staff throughout the NAC are numer-ous but ineffective. Over many years, a lack offunds for operations, absence of effective pro-grams, outmoded procedures, and ineffectivemanagement have demoralized the public sec-tor workforce.6

Overall, the AKRSP operates in a strongly sup-portive environment, drawing on a fund of gov-ernment goodwill based on its performance (Joint

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

6

To assess the challenges of the AKRSP it is necessary to under-stand that the Northern Areas (although not Chitral, which is partof North West Frontier Province) are militarily sensitive andpolitically different from the rest of Pakistan. For historical rea-sons it is a federally administered area under the jurisdiction ofPakistan’s Federal Minister of Kashmir Affairs and Northern

Areas and a Legislative Council, headed by a federally appointedchief executive and consisting of 24 locally elected memberswho elect a deputy chief executive. The people of the NorthernAreas do not elect members to Pakistan’s National Assembly, norare they directly taxed, but they are provided a number ofsubsidies.

U n i q u e P o l i t i c a l S t a t u s o f t h e N o r t h e r nA r e a s

B o x 2 . 1

Review Mission 1999, p. 76). But in parts of theNAC government establishment, the AKRSP isregarded with a mixture of admiration, envy, andeven suspicion, usually because of its roots in theIsmaili community. This would be a handicap ifit were not for the positive view of the AKRSP heldby the most senior government officials who, rec-ognizing their own limited capacity, see the AKRSPas complementing government activities.7 TheAKRSP also commands respect and admiration ingovernment circles in Islamabad, where the intel-lectual debt owed to the AKRSP and its foundersis freely acknowledged at the very highest levels.Throughout government, however, there is agrowing realization that it can no longer effectivelyabdicate its responsibility in the NAC to the AKRSPand the AKDN, that it must also become moreresponsive to the demands of communities. Thischange of circumstances is creating new oppor-tunities and challenges for the AKRSP in its rela-tionship with government.

EfficacyEfficacy of the program is rated fully satisfactory,in most respects highly satisfactory, but therehave been some areas of weakness. (The extentto which the AKRSP has been an effective insti-tution—a connected but different question—iscovered later under the section on institutionaldevelopment.) For physical achievements, seeAnnex J, tables J.1 to J.11.

The main objectives over the program period,expressed in evolving but largely consistentobjectives or mission statements, can be sum-marized as: increasing living standards andincomes, enhancing people’s capacity through thedevelopment of skills and community organiza-tions, the provision of economic growth oppor-tunity, program replicability, and the oft-quoted“working itself out of a job” objective. How doesthe program rate against these objectives?

With respect to living standards and incomesobjectives, average farm household incomeshave certainly more than doubled. In real terms,between 1991 and 1997 alone they more thandoubled, except in the more challenging Astorearea. While some of these gains may not havebeen sustained, and while the trend probablywas not so strongly evident in the pre-1991

period, it seems clear that objectives for income,and probably living standards as well, wereachieved. The extent to which this is attributa-ble to the AKRSP is less clear. While the origi-nal objectives did not claim that the anticipatedincome increases must be wholly attributable tothe AKRSP (and there was probably an implicitassumption that some would not be), there issome evidence from the AKRSP socio-economicsurveys that about one-third is attributable,although there remain methodological prob-lems. Based on the economic analysis, the causallinkages are probably mainly through impacts ofinfrastructure and natural resource management(NRM) interventions on net farm incomes. Inqualitative support of this finding, increasedincomes substantially attributable to the AKRSPand greatly increased land values from infra-structure development were widely noted by vil-lagers during mission field visits.

With respect to the enhancing capacity objec-tives, the program appears to have been very effec-tive for men, but less so for women. Generallyimpressive community organizations have beencreated, at least for men (Annex J, tables J.1, J.3,and J.4), and there has been a substantial levelof training of individuals in a wide range of skills(Annex J, tables J.10 and J.11). Village capacityto manage affairs and command resources iswidely acknowledged within the villages to havebeen substantially enhanced by AKRSP support.

With respect to the objective of economicgrowth, progress has been quite strong in agri-culture, livestock, and forestry, including thecreation of about 48,000 hectares of new cul-tivable land—about a one-third increase in cul-tivable land area—and intensification ofproduction on existing land. (Annex J, table J.5,but note some slowing in recent years, partly dueto previous adoption.) But there has been lessprogress outside agriculture. However, it isapparent that the AKRSP’s original mandate wasmainly in the rural economy, especially agri-cultural production, such that any reach by theprogram beyond rural-based production maybe reasonably considered a bonus.

With respect to the objective of replicability,achievement has been very impressive, withmany programs within Pakistan—including the

O v e r a l l P r o g r a m A s s e s s m e n t

7

large National Rural Support Program (NRSP) andthe current World Bank–supported Poverty Alle-viation Project—drawing their inspiration anddesign substantially from the AKRSP experi-ence. (The issue of replicability is discussedmore fully below.)

With respect to the objective of “workingitself out of a job,” efficacy has been modest tonegligible, but the relevance of this finding isquestionable. The evaluation team supports theconclusions of some earlier evaluations thatquestion the realism of that original objective,at least over the period proposed. Moreover, theobjective was later dropped, or at least playeddown. Nevertheless, there have been weak-nesses in the extent to which the AKRSP haspursued more differentiated support at the com-munity level that would have used resourcesmore efficiently.

Efficiency

Economic Rate of Return The economic analysis was undertaken jointlywith the AKRSP, but with the evaluation teamcommenting on and, where necessary, adjust-ing methodology and reviewing assumptions.The benefits of the AKRSP program are partic-ularly difficult to quantify because of the rangeof benefit types and the difficulty of quantify-ing many of them. They include benefits fromthe following (some of which, if added together,would represent degrees of double-counting):infrastructure; natural resource management,including input supply; enterprise develop-ment; benefits for other local programs draw-ing on AKRSP experience, such as the NRSP;training impacts beyond those reflected directlyin such interventions as infrastructure and agri-culture; lowering the operating costs andimproving efficacy for other AKDN institutions;benefits accrued by or through government, forexample, in the recent support for training forDehi Councils; benefits from microfinance—some reflected in natural resource manage-ment benefits, some lying elsewhere, as inhealth benefits arising from improved diets andconsumption loans; reduced loss of life orimproved health through improved road access

to hospitals; positive environmental externali-ties, for example through re-afforestation, butsome environmental negatives also; benefitsfrom marketing support in improved prices;benefits in the area of non-farm employment;support to government in animal disease out-breaks; support related to tourism and bio-diversity, such as the initial support for theIUCN program; broader institutional impacts,such as increasing service capacity for hydelsin Chitral and growth of the related hydel man-ufacturing and service industries; and, moregenerally, increased social capital, some reflectedin benefits listed above, and others additional.

The comparatively high per-household costof the program appears to be justified by the ben-efits. Quantifying within the list above only theinfrastructure and natural resource managementbenefits (which include NRM input-generatedbenefits triggered by microfinance), the esti-mated economic rate of return (ERR) frominvestments between 1982 and 1999 calculatedby the mission is in the range of 16 to 24 per-cent (Annex H). However, alternative and veryplausible scenarios go much higher, and thetrue ERR would be greater in any scenario if allunquantifiable benefits could be added. Briefly,the benefit stream leading to the 16 percent isconservative and is obtained by taking theAKRSP’s worst-case estimated net benefit streamsfor infrastructure, omitting one high outlier fromthe case studies in each infrastructure type,adding the AKRSP’s natural resource manage-ment net benefit estimate, but reducing that by33 percent,8 and also attributing 5 percent of theestimated incremental non-farm income overthe period to AKRSP interventions. These netbenefits are then set against the AKRSP’s totalcosts.

The purpose of taking the worst-case infra-structure scenario and reducing the NRM streamwas to adjust somewhat for the optimisticassumptions identified in the AKRSP estimates(although the NRM productivity coefficientswere based on Food and Agriculture Organiza-tion [FAO] estimates) but, more important, toallow for some “without project” benefits. The“without project” scenario is particularly difficultto model in this case. First, the area, being a

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

8

desert, has virtually zero production when notirrigated. Second, NRM inputs such as theimproved varieties provided through the AKRSPare unlikely to have been obtained in significantquantities through any other means.

Estimating the benefit stream using a differ-ent approach, through estimates of aggregateincome increases over the program period andattributing 10 percent of net farm incomeincreases to the AKRSP, plus the same 5 percentof non-farm income, gives a much higher ERRof 33 percent. The analysis suggests that oneneeds to attribute only about 9 percent of theaverage estimated incremental net farm benefitsof participating households (excluding off-farmbenefits) to the AKRSP up to the year 2010 toachieve an ERR that approximately equals theopportunity cost of capital. However, this eco-nomic analysis should be regarded as partial andindicative, and interpreted with great care, eventhough it is based on survey evidence, infra-structure case studies, other field investigations,and technology innovation models.

The NRM approach captures benefits onlyfrom infrastructure and natural resource man-agement and there are a number of uncertain-ties about data and assumptions. With respectto irrigation infrastructure, there is some com-fort that the models may not be too optimisticbecause the mission obtained estimates fromfarmers for increases in land values from irri-gation that were significantly higher in realterms than those found by the AKRSP casestudies. Moreover, these capitalized land valueswere found to match the projected productionnet benefit streams quite well, suggesting thatfarmers have a good appreciation of the valueof land in relation to future income-generatingcapacity.

The lower 16 percent ERR case is moderatelysensitive to cost and benefit changes, but, asnoted above, this case is conservatively esti-mated. In the higher ERR scenarios, in the rangeof 24 to 33 percent, sensitivity is lower. ERRs inthese ranges are well above the opportunitycost of capital—assumed, for Pakistan, to be 12percent. (As a comparison, the average for WorldBank–financed agriculture projects, where suchanalysis was attempted, was 22 percent over the

1996 to 2000 period, but very few of thesewould have been in such a challenging area.)

The Problem of Attribution Notwithstanding the indicative economic analy-sis that provides part of the efficiency story,attribution of the AKRSP’s impact on overallgrowth is still difficult to establish through theanalysis of survey data. The AKRSP has nowspread to almost every corner of the NAC, rul-ing out the collection of comparable “with andwithout” data. Benchmark studies were not con-ducted before the AKRSP began work. Moreover,even if evidence of change were available, itwould still be necessary to separate the effectsof the AKRSP from those of government andother players. These cannot be lightly discounted,as they include the construction of the Karako-ram Highway and substantial commodity andpower subsidies, as well as the consequences ofextensive military investment.

Nevertheless, there is much anecdotal evi-dence of the economic benefits of AKRSP pro-grams. Villagers often estimated that in the pastfive years, their incomes had increased (by 10percent, 20 percent—and in some cases by 50percent) due to AKRSP interventions. One groupof villagers explained that deaths, especiallydeaths of women in childbirth, which had beenparticularly high in their area, have been muchreduced by the completion of their access road.Now they can get to the hospital in two hoursrather than two days.

Evidence of substantial increases in per capitaincomes and expenditure in the NAC comesfrom household surveys conducted by theAKRSP’s Monitoring and Evaluation Section.Between 1991 and 1997, the latest year of sur-vey, real incomes per capita rose by about 2.6times, although the changes before and after thisperiod were probably much more modest. But,for the reasons noted earlier, income changescannot be conclusively or wholly attributed toAKRSP interventions.

In an attempt to explore this issue, a modestamount of further analysis was carried out by amission consultant on the 1991 and 1997 AKRSPfarm household income expenditure surveys.9

The surveys were found to be generally of high

O v e r a l l P r o g r a m A s s e s s m e n t

9

quality—an evaluative finding in itself. The con-sultant was able to replicate many of the resultsfrom the data files. Strengths include the extentof detail and level of disaggregation in the col-lection of data on costs, output, and values bycrops. Weaknesses include the relatively smallnumber of questions on non-farm income, whichrepresents a substantial share of total income,and lack of recording of consumption from ownproduction. There were some problems in track-ing the coding. Main findings, which were quiterobust, include the following:• With respect to attribution, while, as noted,

there are almost insuperable difficulties inthe absence of “with” and “without” AKRSPsupport and “before” and “after” AKRSPcontrol areas, the data suggests that totalearnings growth in villages with village organ-izations is about one-third higher than thosewithout, possibly with a stronger effect onland-poor households than land-rich house-holds. This last finding, however, was gen-erally true in percentage terms but not alwaysin absolute terms, and the differences wereonly weakly significant. However, there wasno significant effect of being a village organ-ization (VO) household per se. This isprobably because—as observed by the mis-sion—benefits tend to spread to all villagehouseholds, not simply VO members. In eco-nomic terms, there are free riders.

• There is not much higher household incomeunder double-cropping than single-cropping,particularly in 1991. This is probably partlybecause land holdings in double-croppedareas are smaller, but also that length of grow-ing season and fertility may be more theexplanatory variables than simply the num-ber of harvestable crops squeezed in.

• The decline found in the surveys in house-hold expenditures is probably not of concern,since it shows a decline in total food expen-ditures, particularly fruit. With the rise in pro-duction, this probably suggests that fruitconsumption has increasingly been comingfrom own production.

• Growth in livestock income appears to be atleast partly attributable to the support pro-vided to the community organizations,

although forestry income growth appears tobe negatively affected by the presence of aVO. This possibly arises from village organ-izations inducing more careful managementof forest resources—observed by the mis-sion—thus lowering short-term forest incomesfor substantial longer-term benefit.

• A major source of income growth, at least forthe land-poor households, is non-farm activity.

• With respect to Astore, an area of particularconcern, exploration of attribution is evenmore difficult. There were no VOs in Astorein the 1991 survey, so comparisons seekingattribution comparing income differentialsbetween VO and non-VO at the early stageof the program with VO and non-VO differ-entials at the later stage are not possible,although the VO/non-VO difference in Astorein 1997 (which may reflect both selectionand treatment) is comparable to that in otherregions. In the absence of good measures ofinfrastructure, it is also difficult to demonstratethe impact of poor infrastructure in Astore,although there is some evidence of lowerinitial fruit income and fruit income growthin Astore, which could partly reflect trans-portation problems.

Recommendations by the consultant for futuresurveys include: better recording of consump-tion from household own stocks, including con-sumption items such as schooling and clothing;recording of gender assignable goods such aswomen’s clothing, which would help to indicateintrahousehold allocation; dates of VO estab-lishment and other local institutions and theirareas of action; dates of provision of publicgoods (roads, schools, health centers, waterresources); and distance to markets. In general,future data collection in both surveys and mon-itoring and evaluation (M&E) warrant moreattention to gender.

Comparative Cost-EffectivenessComparing the AKRSP’s costs with those ofsimilar organizations reveals another aspect ofefficiency. The relevant data, summarized inAnnex C, show that the AKRSP incurs totalcosts per beneficiary that, on a 5-year basis,

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

1 0

fall within the range of costs incurred by com-parable projects. AKRSP total costs per house-hold, including both investment and operatingcosts, are almost exactly the same as those ofthe average for the International Fund for Agri-cultural Development (IFAD), which special-izes in supporting community-based programsfor the rural poor. However, the AKRSP has beenoperating for a much longer period than mostdonor-funded projects. Consequently, on a full-program-period basis, regardless of programlength, total costs per beneficiary are high com-pared with typically shorter programs.

While it might seem justified to consider thetime variable on the grounds that, for equity rea-sons, poor households should only receive a cer-tain lifetime level of support, it could also beargued that very few 5-year donor projects rep-resent the only external support that thosehouseholds have received over, say, a 20-yearperiod. Operating costs are much more difficultto compare. As concluded in a recent World

Bank OED study of social funds (World Bank2002), comparable data are available in very fewcases, given differences in benefits. However,at about US$40 per household annually, theAKRSP is at the top end of the range.

Cost Trends Over TimeThe balance of costs has changed over time, asshown in figure 2.1, but it is not easy to drawefficiency conclusions from this. Operating costas a percentage of total cost has risen margin-ally, from a little below 30 percent in the firsttwo years to a little above 35 percent in the mostrecent two years, but with a peak of nearly 50percent in 1991. (Operating costs, as definedhere, include capital costs for operation otherthan the program investments, salaries and con-sultancies, travel, administration, office mainte-nance, and vehicle operation. Total costs, asused in figure 2.1, include the disbursement fig-ures for credit rather than the net credit fig-ures.) The largest jump in costs was very early

O v e r a l l P r o g r a m A s s e s s m e n t

1 1

E x p e n d i t u r e C a t e g o r i e s a s aP e r c e n t a g e o f T o t a lF i g u r e 2 . 1

◆

◆ ◆◆

◆

◆◆

◆

◆◆ ◆

◆ ◆ ◆

◆

◆

◆

◆

■■ ■

■ ■

■ ■■

■ ■

■

■

■■

■■

■

■

▲▲

▲▲ ▲

▲

▲ ▲ ▲ ▲ ▲ ▲ ▲ ▲ ▲ ▲ ▲ ▲

✽✽ ✽

✽

✽

✽ ✽

✽ ✽ ✽ ✽ ✽ ✽✽ ✽

✽✽

✽

✳ ✳

✳

✳

✳✳ ✳

✳

✳✳

✳

✳ ✳ ✳✳

✳ ✳

✳

1982–83 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000

0

10

20

30

40

50

Perc

ent o

f tot

al e

xpen

ditu

re

◆ Operating cost

■ Salaries/consultancies ▲ Training

✽ Research/survey/demonstration

✳ Infrastructure

Year

in the program, from 1983 to about 1988, withan almost flat trend since that time (Annex J, tableJ.15). This increase occurred at a time of sub-stantial program expansion, not only with respectto number of villages, but also to breadth of com-ponent coverage. The growth from the approx-imately 15,000 households served in 1982/83 tothe approximately 95,000 served currently hasinvolved going into increasingly challengingand remote locations. The substantial operatingcost increase in 1999 is attributable to the pur-chase of all the vehicles for the current phasein that one year. Operating costs per house-hold, in constant dollars, have been relatively flatsince 1988. Infrastructure as a percentage oftotal cost has fallen from about 50 to about 20percent. Salaries as a percentage of operating costhave risen. Training as a percentage of totalcosts has remained fairly steady and research,survey, and demonstration declined until about1998, then rose again to about the current 15percent.

The increase in operating cost as a percentof total peaked in 1991, while the share of infra-structure declined; this could be interpreted asa period of relative inefficiency. The improve-ment after 1991 could be interpreted as a periodof improved efficiency. But interpreting effi-ciency from operating cost percentages in thistype of an organization can be misleading. First,an organization with a focus on communitymobilization and capacity building (as opposedto investment activities) could be deemed highlyefficient even when operating costs are 100 per-cent of total costs, since all resources would befor operational activity. Second, although theAKRSP seems to be at the top of the range ofcosts per household among its comparators, theERR justifies that cost. With respect to subsidycosts, NRM subsidies have declined significantlyand there is now more focus on increasing vil-lage contributions in the infrastructure program.

Sectarian Allocation of InvestmentsAKRSP data indicate that the pattern of supportto date for the approximately 73 percent non-Ismaili population has been 67 percent of theorganizations (village organizations and women’sorganizations) formed, 66 percent of infrastruc-

ture projects funded, and 50 percent of credit dis-bursed—the latter based on effective demandwith no restrictions. Given the early focus onGilgit, which has a larger share of Ismaili pop-ulation than Chitral or Baltistan (which has vir-tually no Ismaili population), and given theinitial reluctance to partner with the AKRSP bysome non-Ismaili villages, the overall balance atthis stage does not seem to represent strong evi-dence of a sectarian bias in meeting demand.Recently, however, the AKRSP board has directedthat, in Chitral—where sectarian resistance to theprogram has been strongest and the investmentto date in non-Ismaili villages has been lowest—two-thirds of infrastructure investments shouldgo to non-Ismaili communities. This will behelpful in improving the balance as well as inaddressing poverty. Based on a categorizationexercise of villages into poor and rich accord-ing to criteria that include land availability, live-stock holdings, off-farm incomes, and access toservices, the percentage of poor Ismaili villagesin the program area is estimated at 45 percentand the percentage of poor non-Ismaili villagesis estimated at 55 percent.

Subsidy Policy and Cost Recovery At a broad level, the whole of the AKRSP’s out-lays constitute a subsidy to the rural economyof the NAC. Thus, and following standard eco-nomic principles justifying the public provisionof economic goods, cost recovery should be akey element in the AKRSP’s development strat-egy and financial plans. Within this broadaccounting, the pattern of targeted subsidiesand cost recovery for infrastructure, for some nat-ural resource inputs, for technical support, andfor training requires rationalization.

The current grant structure appears to haveevolved rather than to have been chosen explic-itly for efficiency, and grants are thus neither eco-nomically optimal nor based on a clear rationale.For example, one might expect to see lowerlevels of subsidy for infrastructure with moreprivate good content, such as irrigation andmicrohydel (small-scale hydropower),10 than forinfrastructure with more public good content,such as roads. Yet the reality is the reverse. Thelevel of grant in 1998/99 averaged about 70 per-

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

1 2

cent for irrigation and about 85 percent for micro-hydel, compared with about 55 percent for roads.

While subsidies for public goods or publicgood elements can be justified, the AKRSP shouldnot generally subsidize private goods. Therewould, however, be a case for such subsidyunder the following circumstances: (i) wherethere are substantial positive economic exter-nalities; (ii) on common property goods wherecommunity welfare is at stake;11 (iii) where inter-ventions reduce poverty, provided they are welltargeted and do not create perverse incentives;and (iv) where there are environmental exter-nalities.12 But criteria need to be developed fortermination when the criteria are satisfied.

A high grant element, and relatively the high-est, on microhydels seems particularly difficultto justify. Economic analysis shows that theseprojects have the highest rates of return. More-over, cost recovery in microhydel is less chal-lenging than with many other types ofinfrastructure. The question is less whether a sub-sidy for microhydels is justified than whetherthose funds could be better used elsewhere,perhaps in the poorer, high-altitude villages forhard to obtain inputs or communications. Inshort, a review of subsidies measured against thefull cost of provision is needed, followed by theformulation of a clear policy.

However, a clearly articulated subsidy policyis only part of a full approach to cost recovery.Many of the AKRSP’s activities beyond the pro-grams noted above are aimed at the provisionof goods that are neither pure public goods noropen-access goods. These quasi-public goods,such as training in villages, can be internalizedand used for private gain. Under those circum-stances, people are often willing to pay at leastpart of the cost of provision. In formulating itsfinancing plans, the AKRSP should also takethis into account, as willingness to pay helps tosignal the amount of such goods the AKRSPshould supply, thereby helping to avoid waste.

In a similar vein, the AKRSP should reviewits entire program to identify all areas where fullor partial cost recovery is possible. This exercisewould help to arrive at a defensible estimate ofthe share of the AKRSP’s costs that cannot legit-imately and practically be recovered and must

be met by other means, such as donor grants orincome from an endowment. With such estimatesin hand, the AKRSP should develop a clearly dif-ferentiated and time-bound program of costrecovery and embed it firmly in its long-termfinancing plan.

Potential Gains in Efficiency from IncreasedCollaboration and Integration in the AKDNThe efficiency of the AKRSP and the AKDNcould be increased by a planned, structured, butgradual process of integration. Although mak-ing better use of physical facilities, plant, offices,and equipment offers potential gains throughreduced overheads, the most important gainswould come from coordinated and integratedplanning and implementation of schemes inwhich inputs are required from different AKDNagencies. Additional benefits from synergy effectsof joint operations are likely to be substantial.

The AKDN has been slow to pursue thisopportunity. At present each agency has sepa-rate offices, staff, vehicles, and equipment andthinks and acts separately.13 “We operate insilos,” said the general manager of one AKDNagency. Yet all depend on the social infrastruc-ture created by the AKRSP, have similar devel-opment goals, operate in overlapping areas,and serve the same people. Generally the peo-ple served do not distinguish between them.Moreover, as incomes rise and communitiesbecome more organized (largely in response toAKRSP-led development), the demand for bet-ter health care, water supply, and educationrises. Other than some informal cooperation inthe field, there is no mechanism within theAKDN that allows a coordinated effort to meetthese demands. The AKRSP has an extensivepresence and is actively engaged with villagewomen—usually key players in matters of healthand education. There is substantial opportunityfor mutually beneficial gains in efficiency.

Institutional ImpactAddressed here are (i) the broader issue of theAKRSP’s institutional impact on the NAC; (ii) thenarrower issue of the effectiveness of the AKRSPas an organization; and (iii) the replicability ofthe program.

O v e r a l l P r o g r a m A s s e s s m e n t

1 3

Overall, the AKRSP is an effective organiza-tion, willing to change and evolve to achieve itsdevelopment goals. (See the timeline in AnnexD showing major transition points by programcomponent.) It learns from the past, reflects onits future, establishes clear objectives, anddeploys skilled staff equipped with adequateresources. It has profited greatly from long-serving and consistently good general managersand key senior staff from its inception. As far aspossible it has relied on recruiting senior stafffrom within the NAC, or has helped NAC staffgrow into more senior positions. This staffing his-tory will stand it in good stead as it confrontsthe challenges of the future.

Institutional Impact of the AKRSP in the NACAs with all institutions, however, questionsremain. The AKRSP’s impact may have come ata broad institutional price. If the AKRSP were tobe wound up tomorrow, there would be a largeinstitutional gap in the NAC, a gap that otheragencies, and most important the district admin-istration, could not fill. While the AKRSP has con-tributed positively to government initiatives inthe NAC, and even sought to influence nationalpolicy, overall, in the district administration, pri-vate sector development, and provincial andnational policymaking forums, it is harder to dis-cern the AKRSP’s handiwork.14 Yet it is successin these arenas that will largely determine thecourse of the region’s future progress.

AKDN organizations and government bodieshave joint interests in responding effectively topeople’s needs. Like the AKRSP, these otherinstitutions have the skills and knowledge to helpbuild public sector capacity. Although it is likelythat government’s presence and program in therural areas of the NAC (especially its agriculturalpresence) were slight before 1982, when theAKRSP was established, its presence and pro-gram are no greater now. Whether this is becausethe AKRSP has been so efficient and effective thatit has displaced government’s programs, givinggovernment little incentive to act on its ownaccount, or whether this is independent of theAKRSP’s presence is difficulty to prove.

Among AKDN agencies the institutional debtto the AKRSP is freely stated. Here, however, the

AKRSP’s influence is to be found in the depend-ence of AKDN agencies on the social capital thatthe AKRSP has created over many years (AnnexJ, tables J.1 to J.5). For example, a rough esti-mate by Aga Khan Health Service suggests thattheir development costs would approximatelydouble in the absence of the village organiza-tions established by the AKRSP. It follows thatsome fraction of the benefits resulting from theactivities of other AKDN agencies is properlyattributable to the AKRSP. It is also reasonableto deduce that the AKRSP’s success in workingwith mixed and non-Ismaili communities helpedto bring about the broadened mandates of otherAKDN institutions.

Among other agencies in the NAC, such as theIUCN, there is a strong appreciation of theAKRSP’s pioneering work in social organiza-tion, which has enabled newer and valleywideenvironmental programs to make a much fasterstart than would otherwise have been the case.

Effectiveness of the AKRSP as an OrganizationOrganizational Restructuring. To address futurechallenges, the AKRSP is rightly proposing toadjust its current structure (see Annex K for cur-rent structure) to align it more closely withadministrative districts. The evaluation teambelieves that these organizational changes mustbe guided by the substantial changes in direc-tion and strategic focus that the AKRSP is beingcalled upon to make, as well as by the new loca-tional realities provided by improved commu-nications through the Internet. This matter isdiscussed in greater detail in the “Future Direc-tions” section.

An Institutional Survey. A survey of profes-sional staff (see Annex E), carried out by the eval-uation mission with substantial AKRSP assistance,reveals the AKRSP to be an effective organiza-tion with a healthy and broadly positive institu-tional climate. This is the first time that such asurvey has been carried out within the AKRSP;partly for that reason, the results need to beinterpreted with caution. Apart from a few weak-nesses noted below, the survey reveals the AKRSPto be an organization staffed by people with aclear sense of purpose, a strong and sharedcommitment to its mission, the competence to

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

1 4

implement its programs, and a commitment tomonitor and learn from their experiences.

There was little variation in response by grade(the survey covered grade 4 and above) orregional office—a strong indication of a highlyhomogeneous institution. Although the AKRSP’sdirection (mission and methods) is well under-stood, the perceived link between personal per-formance and institutional goals shows somerelative weakness. Staff clearly believe that theAKRSP has high standards of openness andequity, and share these values themselves. How-ever, there are signs that these values areunevenly practiced across the organization, sug-gesting the need for some improvement in per-formance. Women were more positive than menon this aspect.

The aggregate response on the humanresource questions was lower than desirable,which indicates that staff think some humanresource policies are unfair and that womenare inadequately represented. The two concernsmay well be interrelated. Similarly, although thegroup of questions on authority, responsibility,and accountability records a high score overall,there are indications of possible unfairness in theapplication of accountability and in the report-ing of bad news. Management should consideran open, focused response to this.

The section on capability and competencereflects well on the AKRSP, but there are con-cerns about coordination and centralization.This may reflect current discussion about the roleand performance of Field Management Units(FMUs; see Organizational Charts, Annex K). Italso probably reflects the tension inherent inspreading the Core Office across two locations,Gilgit and Islamabad. There is some concern evi-dent about the AKRSP’s tendency toward intro-spection, but otherwise the AKRSP is clearlyseen by staff to be a learning organization thatis open and outward-facing.

Human Resource Development. As the resultsof the institutional survey suggest, the AKRSP hasa competent and generally motivated staff. TheAKRSP recruits actively and offers competitiveremuneration.15 It also maintains close links withseveral overseas academic institutions for bothtraining and advisory purposes.16 The AKRSP

invests heavily in training. As a result, professionalstaff spend a high proportion of time in training—just over 10 percent in the Monitoring, Evaluation,and Research Section, for example. Consultants,technical assistance, and high-quality volunteersare used judiciously. Staff turnover is modestdespite the remote location.

The proportion of women among the AKRSP’sprofessional staff (grade 4 and above) is about14 percent, well below the AKRSP’s target of 30percent by 2003. For an organization that placesa high priority on gender equity, this is an unsat-isfactorily low figure, suggesting that the AKRSP’sgender policy is not being pursued aggres-sively.17 Implementation of the gender objectivesmay also be uneven.18 Recruitment of womenusing conventional passive techniques is diffi-cult and slow. Active recruiting and other tech-niques need to be used, and there must be awillingness to take risks, including hiring qual-ified women from “down-country.”

There are no women in the core team and noone (male or female) in the core team is respon-sible for the women’s program. A woman shouldbe appointed.19 The present lack of leadershipleaves female staff with no one to talk to aboutissues they face. Some women also feel that theirtraining and professional specialization is notbeing used, and they are there to support thewomen’s program only. Last, there seems to bea reluctance to have men working for thewomen’s program. Most organizations haveroughly equal numbers of men and womenworking on gender issues. While such a ratiomay still be unrealistic in Pakistan, greater exper-imentation is in order.

Funding, Financial Management, and theBudget. The AKRSP’s annual expenditure hasgrown from Rs. 14.6 million in 1983 to Rs. 346.8million in 2001. The donor share, excluding theAga Khan Foundation, has grown from about 30percent to 95 percent. The increasing contribu-tion of donors has allowed the AKRSP to expandand deepen its program more quickly thanwould otherwise have been the case, but it hasalso created a risky dependency on externalfunding. Most donor programs are due to ter-minate by the end of 2002 unless new agree-ments are reached.

O v e r a l l P r o g r a m A s s e s s m e n t

1 5

The AKRSP has a well-developed and well-managed budget process targeted at the fulfill-ment of its principal programs. The annualbudget cycle begins in August with the issuanceof budget guidelines to Field ManagementUnits—the first level of accounting. From therethe budget is discussed and built up through theregions and Core Offices into a consolidatedwhole, ready for the start of the ensuing fiscalyear. Once framed, a comprehensive budgetdocument is presented for board discussion.This budget process is complicated, however, bythe need to fully integrate, yet separately accountfor, donor funds.

The chart of accounts needs to be thoroughlyrevamped to reflect the AKRSP’s logframe ofinputs, outputs, and outcomes. The annualbudget document and the process by which itis generated is a major opportunity to adjust strat-egy in light of events and to plan tactics for thecoming year. But this cannot be easily doneunless the budget is framed and analyzed interms of outcomes and results. At present thebudget structure and analysis, although verycompetently done, is geared to expenditureanalysis of programs and sections, reflecting anexcessive concern with inputs rather than results-based management outputs.

Monitoring and Evaluation (M&E). The Pol-icy and Research Section handling M&E is wellmanaged, well staffed, valuable, and effective,but less so than it might be. Its reports areimportant instruments of learning within theorganization and command a wide readership.The AKRSP has succeeded in hiring young, well-qualified social scientists, supplemented by con-sultants as needed. The unit occupies a keyposition in the Core Office, contributes sub-stantively to strategy and policy formulation,and its manager is a trusted lieutenant of the gen-eral manager. Overall, the quality of the M&Ework is good. As noted earlier, an independentU.S. researcher from a reputable universityjudged the quality of the data analysis in the 1997Socio-economic Survey to be generally good. Inthe view of the evaluation team, however, someof the work lacks analytical depth and, likealmost all similar units, has failed to grapple withthe fundamental problem of attribution. There

should also be more attention to gathering datarelevant to the gender program.

The M&E program should embrace moreanalytical tasks and experiment with newer eval-uative methods such as beneficiary assessmentand participatory evaluation. To achieve theseaims, the M&E work program will need somerestructuring to release resources. At present,progress and related reports for donors imposea heavy burden with high opportunity costs.Donors should simplify their informationdemands and agree on a single annual report.Such a step would free resources for more inno-vative M&E.

The proposed Policy Unit would lead to otherchanges in M&E. A sharper separation of mon-itoring from research and evaluation is sug-gested for the unit. Reflective and analyticaltasks would be confined to the new unit, whilemonitoring in the field would take on more ofa “progress-chasing” role. However, location isimportant. Experience in other countries suggeststhat the relevance of evaluation, research, andpolicy work in rural development declines as itscontact with developments in the field weakens.

The Replicability of the AKRSPThe evidence suggests that the AKRSP is replic-able, but the parts more than the whole. AnnexJ, table J.17 lists eight rural support projects thatto some extent owe their basic approach and(sometimes) key staff to the AKRSP. These proj-ects had established some 20,000 communityorganizations by 2000, about 5 times the num-ber of AKRSP organizations. These conceptualsuccessors are now active in nearly 50 admin-istrative districts. AKRSP experience and skillshave been used in the National Rural SupportProgram (NRSP),20 the IFAD-supported ChitralAgricultural Development Project, KhushhaliBank, recent IUCN interventions, Sarhad RuralSupport Corporation (SRSC), the Ghazi BarothaTaragiati Idara (GBTI), the South Asia PovertyAlleviation Program (SAPAP), the Punjab RuralSupport Program (PRSP),21 and the activities ofother AKDN organizations. The design of thegovernment’s new national Poverty AlleviationProgram is also based on AKRSP experience. Sev-eral other programs and projects are less closely

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

1 6

modeled on the AKRSP but draw on its basicconcepts. Staff trained at the AKRSP have movedon to many parts of rural development in Pak-istan and beyond (see Annex A). Overseas,apart from similar Aga Khan Foundation ruralprograms in India, Kenya, Mozambique, andTajikistan, AKRSP ideas have spread far andinfluenced the design of rural development onnearly every continent. None of the programsmentioned offer comprehensive economic analy-ses that attempt any attribution to the AKRSP. Anevaluation of the NRSP, conducted in 1998, islargely positive in its assessment, but not quan-titative (UNDP 1998). To the extent that theseschemes are successful, some part of their ben-efits is attributable to the AKRSP. Attributingonly a modest share of these benefits—were thedata available—would greatly increase the rateof return to the investment in the AKRSP.

SustainabilityThe AKRSP’s sustainability can be evaluatedalong two main dimensions: institutional sus-tainability and financial sustainability. (Theimportant broader institutional sustainabilityissue related to the vulnerability of the NAC toan AKRSP withdrawal was discussed earlier.)

Institutional SustainabilityThe policy and development ideas that theAKRSP has developed are seemingly sustainableas they are applied and modified in an organicfashion elsewhere. In the NAC, many village insti-tutions and their works fathered by the AKRSPhave already matured and look set to have longlives, surviving independently of their progen-itor, continuing to fill a void left by the declineof earlier political and social structures. Infra-structure works are likely to be sustained becausethey yield a substantial positive return and arewell maintained. In this narrow sense, the AKRSPcan be said to be institutionally sustainable.

Financial SustainabilityThe AKRSP is not financially self-sustaining. Asa private corporation, the AKRSP cannot tax the

people it serves. As a non-profit organization, itcannot trade for profit, and full cost recoverythrough commissions and service fees is notyet feasible, although significant moves in thisdirection can be made. From the beginning, theAKRSP has depended on funding by others, asintended. Table J.14 in Annex J shows the sub-stantial annual and cumulative contributionsmade by international donors, including theAga Khan Foundation (AKF), originally thelargest contributor. But that has changed. Cumu-latively the AKF has now contributed about 13percent of the total. By 2001 its share of theannual budget was down to 5 percent.

Major donors, although gratified to have beenassociated with a successful venture, have beenshowing signs of fatigue, although recent eventsmay alter this. If nothing is done, the AKRSP maysoon face financial problems. Were this to hap-pen, the AKRSP would be obliged to reducestaffing and the depth and range of its work. Theextent to which other donors are waiting in thewings is unclear.

A task force examined the AKRSP’s sustain-ability in 2000 and concluded that the programcould not expect any increase in funding fromits current six bilateral donors, but that othersources (bilateral, international, foundation, andprivate philanthropy) could be developed (TaskForce on Sustainability 2000, p. 33). It also con-cluded that the AKRSP’s financial future wouldbe very tight, requiring it to strenghten finan-cial management and increase cost recoveryand fee income. It also proposed the estab-lishment of an endowment fund that wouldyield sufficient income to cover at least somecore costs.22

But a narrow concern with self-financing maybe misplaced. If the AKRSP tightens its belt, if itdevelops a new strategy pursuing efficiency, ifit develops partnerships in a development coali-tion, and if it sells itself more as a laboratory fordeveloping and testing creative new ideas onrural development,23 then these public goodaspects of its work constitute a strong case forcontinued international donor assistance.

O v e r a l l P r o g r a m A s s e s s m e n t

1 7

1 9

Community Organizations

Community organizations lie at the heart of the AKRSP approach. Theirachievements have been remarkable and, unlike many other donor-funded interventions, sustained. This success has been widely repli-

cated in Pakistan, and, as noted, similar organizations elsewhere haveemulated the AKRSP model. Two main types of organizations are currentlybeing supported—village organizations for men (VOs) and women’sorganizations (WOs) for women, although umbrella local developmentorganizations (LDOs) are becoming increasingly important.

Annex J, table J.3, shows steady growth, to 85percent of rural households with communityorganization membership by 2000, with thehighest percentage in Gilgit (94 percent) and thelowest in Chitral (64 percent), the latter duepartly to earlier reluctance to join the program.Within the villages there is widespread appre-ciation for what these organizational structureshave done for members, and there is inde-pendent evidence that membership brings ben-efits. As one observer noted, “AKRSP hasprovided a platform.” Many villagers’ commentssuggest that a main benefit is that it has empow-ered them to think and make decisions ratherthan simply receive. As one villager put it,“Before, people said, ‘Yes Sir;’ now people say,‘Why Sir?’” However, the extent of ethnographicinformation describing internal village differ-ences and patterns is still quite limited. This

section assesses the relevance, efficacy, and effi-ciency of community organizations, with a par-ticular focus on women’s organizations becausethey appear to be in greatest need of redirec-tion and offer the greatest potential for gain.

Relevance

Over the Period of the ProgramThe VOs have been highly relevant institutions.Fostered by the AKRSP from the outset in 1982,they built on local traditions of self-help andcooperation. For villagers they also proved aneffective channel for accessing AKRSP assis-tance. From the perspective of the AKRSP, theorganizations successfully embodied the dualobjectives of the program: to build village insti-tutions based on participation and to bring abouteconomic development through Productive Phys-

33

ical Infrastructure (PPI). The organizations, par-ticularly the men’s, met both AKRSP and villageraspirations and formed an effective instrumentfor dialogue between male representatives of vil-lage communities and program staff.

Eighteen years later, many villages continueto have active organizations. In others they areless active, or in some cases dormant. In mostareas VOs now coexist with a plethora of sub-and supra-village organizations, including LDOs,cluster organizations, and Social Welfare Soci-eties. Some of these have arisen out of the needto coordinate mohallah-based (hamlets) VOs.Others have developed independently inresponse to specific functional interests or otherfunding sources.

Relevance TodayWhile it is difficult to assess the relevance of com-munity organizations from brief visits, dataconfirm the continued relevance of many organ-izations. The 1998 Institutional Maturity IndexStudy estimates that roughly two-thirds of malemembers attend VO meetings regularly andslightly more than two-thirds of women attendWO meetings. This is a positive indication of rel-evance, as time is a highly valued resource inpoor households. However, based on its fieldinterviews, the evaluation team found reportedmeeting attendance quite variable. Where VOshave not retained their relevance to members andhave disbanded, this may not be an indicationof failure in institution building. It may be a pos-itive indication that villagers have sufficientmaturity and confidence in their improved socialcapital status to feel comfortable letting go of aparticular organization that has outlived its use-fulness—perhaps to spawn another one in adifferent guise, with a different objective.

Maintaining Relevance in the Future VOs will continue to be highly relevant in somevillages—for example, those in single-cropping,high-altitude zones isolated from markets andother funding sources, and where subsistenceagriculture is still the norm, or where the AKRSPplans to offer further infrastructural support.However, the relevance of the original formu-lation of broad VOs may decline in villages as

more diverse and outward-looking livelihoodoptions and new institutions such as function-based farmer interest groups take their place, butoften with some of the same membership. Eco-nomic options have now expanded in the NAC,and the need to build alliances with down-country merchants will be more important. Thisis what is happening now, and what the AKRSPis planning for with its “functions-based”approach. To better understand this processand enable creative responses, the AKRSP needsto assess why some villages have more matureinstitutions than others. The evaluation teamdoes not feel that the 1998 Institutional MaturityStudy really achieved this.1

Relevance of Women’s OrganizationsWith few exceptions, all institutional diversifi-cation at the village level has been achieved bymen in men-only groups. During evaluationteam visits, not a single women’s group planneda presentation; no woman got to her feet to makeeven the shortest speech on behalf of her organ-ization. The contrast between the well-organizedmeetings held by the VOs and the meetingsheld with WOs could not have been greater.Whatever the cultural and religious justificationsfor the enormous discrepancy in women’s abil-ity to plan and manage activities, the twin-trackapproach has created an institutional processwhere one track (the men’s organizations) hasleft the other track (the WOs) far behind.

This is not to suggest that almost 400 women’sorganizations have not been relevant to women(Annex J, table J.5). The problem, perhaps, is thatthey have been too relevant. They haveresponded to women’s practical needs but notto their aspirations or potential as public citizens.The AKRSP may not have set its sights highenough in encouraging women to graduallyincrease their public profile and, with the sup-port of men, create their own institutions, or inthe right circumstances, join men in theirs.2

Relevance of Community Organizations for the PoorestThe relevance of community organizations tothe poorest, who have never been members ofthem, appears to have been much less than for

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

2 0

their generally better-off participating neighbors.Many of these poorest are the destitute—widows,people with disabilities, the mentally handi-capped, and immigrants with few resources.With little time to participate in collaborativework or with insufficient financial capital tohonor reciprocal obligations, they are not easilydrawn into the fold. Experience from elsewheresuggests that these groups often tend tobe excluded. It has taken a long time for theAKRSP to address this social exclusion.3 Non-participating poor families have still benefitedsomewhat from the ability of the village organ-ization to manage infrastructure and naturalresources on behalf of the village. Traditions ofequitably sharing out new land parcels havegreatly helped equity. But as non-members, theyhave missed out in several ways. First, they willhave had no opportunity to save and build upcapital as a social safety net, or to use their col-lateral for borrowing in times of crisis, medicalemergencies, or food scarcity. Second, they willhave missed out on the development of socialorganization skills.

Relevance of the Pilot Poverty ProgramThe pilot poverty program—to test out moreintensive, directed poverty approaches—hasbeen operating for only one year, and so far injust a handful of villages. It was initiated in 1999in both Gilgit and Baltistan as an outcome of theTraining and Learning Program in Social Devel-opment, which has helped staff develop ana-lytical skills.4 It is now enabling the pooresthouseholds to join VOs. This effort will need tobe monitored. In one village (Gakuch Balla) itwas not clear to the evaluation team that the tar-geted families had really wanted to join a com-munity organization or that their better-offneighbors had welcomed them. Experience fromother countries suggests that incorporating thepoorest into well-established village organizationsis very difficult, particularly if poverty is com-pounded by any form of disability (which is thecase for some families in both the pilot villagesvisited). This is a reminder that high levels ofsocial capital within a group are good for thegroup, but can make it exclusive and intolerantof outsiders who are noticeably different.5

Efficacy

Overall Efficacy of Village OrganizationsOverall, VOs have been highly effective. Theyhave been the mechanism for developing socialcapital, creating infrastructure, carrying out sav-ings and loan activities, enhancing naturalresource management, and developing humancapital. However, for the future, resource con-straints will dictate an increasing degree of dif-ferentiation among types of villages andorganizations. Broad strategic questions relatedto differentiation—discussed further in Chapter5—include: Should the AKRSP support high-fliers with good business skills? Or should itsupport organizations that have primarily socialobjectives, such as managing conflict resolutionor canvassing for girls’ education? Or shouldthe AKRSP primarily support economically dis-advantaged areas and villages? Designing aneffective differentiated approach will call for agood understanding of the differences in villagematurity, the reasons for those differences, andthe status of social capital.

Overall Efficacy of Women’s OrganizationsThe evaluation team focused particularly onunderstanding how members of WOs judged“effectiveness.” Not unexpectedly, womenappear to judge an activity to be effective if theythink the time they put into it is justified by whatthey got out of it. The outcome might bebroadly economic, or it might be social. Usingpeople’s own criteria for assessing effectivenessis difficult, but it needs to be tried. It stops out-siders from judging effectiveness only fromtheir own perspective and ensures that the def-initions of participants are also taken intoaccount.6

Most WOs are seen as effective by theirmembers. There is no doubt that for manywomen their organization is an important partof their lives—probably to a much greater extentthan the VO now is for men. For one thing, itis still the only organization for women in themajority of villages, and their first and onlyexperience of self-management. In most vil-lages, WOs appear to provide sufficient socialand economic returns for members to meet

C o m m u n i t y O r g a n i z a t i o n s

2 1

regularly, usually weekly. The AKRSP is appre-ciated for the many opportunities it has providedfor women through these organizations. In gen-eral, there was a “feel good” factor within well-established WOs where AKRSP support wasseen to be making things happen. The conversewas also true. In villages where the AKRSP hasonly very occasional contact because of religiousopposition, debt default on the part of a VO, orwhere visits by women’s social organizers(WSOs) were very infrequent, there was oftena feeling of abandonment. WOs continue to seethemselves as dependent on AKRSP social andfinancial support. The relatively weak per-formance of regional women’s programs hasbeen noted in recent donor reviews and actionsproposed.7

Training Programs for WomenThe AKRSP’s impact studies indicate that train-ing programs through WOs have not alwaysbeen effective in achieving significant or sus-tainable increases in incomes, notwithstandingthe importance of basic numeracy and literacy.Training programs appear to work best whenthey add value to a product such as poultry rear-ing or vegetable production.8

Tailoring classes are common, but incomereturns are modest. Training appears to haveflooded the market with medium-qualitywomen’s garments with a limited village market,because everyone can now sew. To have anysustained financial impact, such training wouldneed to be of sufficient duration to allow womento produce high-quality products (such as men’swear) that are in demand.

Good financial advice is essential, and creditfor start-up capital, perhaps with some grantelement for the poor, would be an additionalboost. Most important, training in marketing(including pricing, procurement of inputs, anddifferent market channels) is essential. Womenneed practical help finding markets and engag-ing middlemen. The Enterprise Section (seeAnnex K, Organizational Chart) provides mar-keting advice, but to individuals and marketinggroups, not to WOs. Learning Support Units(LSUs) therefore organize training in a “demand”vacuum.

AKRSP Social Sector Interventions and StaffingSocial service programs channeled through WOsappear to be only modestly effective. Two exam-ples, water supply and women’s literacy, areaddressed below.

Water supply schemes and community bath-rooms are much in demand by WOs. Such pro-grams are now being channeled through them,which indicates that the AKRSP clearly appre-ciates the value of making the WOs the conduitfor village services. It will be important to ensurethat women will manage these schemes andthat men will not take over. The evaluationteam was concerned about whether the AKRSPis organized and staffed to invest the timerequired at the village level to ensure this out-come. Men need to be to helped to appreciatethe reasons for the shift in policy, to support it,and to make the connection between women’smanagement of small-scale infrastructure and vil-lage development as a whole. The lack ofwomen staff in technical departments is likelyto be a constraint to their development of tech-nical skills.

The scale of support for women’s literacy isimpressive—48 women’s literacy centers wereset up during the year 2000. The AKRSP is pur-suing two models. One involves hiring existing,often male, teachers to provide classes forwomen and uses standard adult literacy tech-niques. The other, in Chitral, is developing andtesting its own methodology and materials, andtraining its own teachers. It is highly participa-tive, with strong functional objectives (materialsaim to change health, hygiene, and nutritionbehavior as well as to teach literacy and numer-acy). The program is of high quality, but alsoreputedly of high cost.

The AKRSP could use the opportunity offeredby these two approaches to explore a funda-mental strategic question and reflect the answerin the strategy: What is the objective of provid-ing literacy classes? Are classes intended to makewomen literate as a stepping-stone to otheractivities, or are they mainly for empowermentand awareness? Literacy, like all components ofthe women’s program, needs to fit into a largerstrategic framework for village and regional(and not just women’s) development.

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

2 2

What Does the Success of CommunityOrganizations Mean for Global Learning about Participation? The success of community organizations in theAKRSP strongly supports other findings by OEDin relation to participation (Blackburn, Chambers,and Gaventa 2000), including the importance ofpersistence; the need to avoid throwing largeamounts of money at a problem over a shortperiod, as in many large donor projects; theneed to avoid having participation “bolted on”to a program design rather than having thedesign subservient to the participation needs; therisk of scaling up too rapidly; the need forNGOs, governments, and donors to adapt pro-cedures to participation; and the need to use“champions” and alliances. One area where theAKRSP has not been as effective is in participa-tory monitoring.

Efficiency

Overall Efficiency of the CommunityDevelopment ActivityThe ability of community organizations to lever-age a range of activities, including infrastructure,microfinance, natural resource management,and capacity building, and also support otherAKDN activities, and that the overall rate ofreturn appears to be satisfactory, suggests thatthe community component has probably beenefficient. However, three particular areas willneed attention to maintain efficiency in thechanging environment: the need to differenti-ate village and organization support—puttingless into the more mature organizations andmore into the less mature and poor organiza-tions; the need to bring women into the main-stream of decisionmaking, while also workingwith the proposed “functions-based” and supra-organizations—there may be some tradeoffhere; and the need to monitor the efficiency ofthe poverty pilot activity, and then make deci-sions about the activity. The last is expanded onbelow.

The Need to Monitor the Efficiency of thePoverty Pilot The AKRSP will need to carefully monitor the effi-ciency of the pilot poverty project, which, whileit has commendable objectives, could result insuboptimal allocation of resources. The pro-gram is well conceived.9 It acknowledges thatpoverty and vulnerability result from a combi-nation of factors: social exclusion, lack of mate-rial assets, and a lack of competencies resultingin limited livelihood options. But the opportu-nity costs of the program are considerable. Oneestimate was that a visit was required to pilot vil-lages once a week for six months, an eight- totenfold increase in reported “normal” frequency.

If the pilot poverty program is to be extended,the AKRSP may want to consider a dedicatedteam of staff seconded to this program. Thosecurrently involved have learned a great deal, andit would be inefficient not to capitalize on theirskills. Alternatively, an NGO could be contractedto oversee this program.

In any event, once any obvious potential man-agement efficiency gains have been exploited, theAKRSP must ask of the monitoring data whetherstaff time is best spent on an expanded, separatepoverty initiative or elsewhere in the program.Testing various levels of AKRSP support in dif-ferent cases would help answer this question byoffering more of a range of support intensity totest. It may be that the learning gained so far couldbe mainstreamed fairly quickly into other pro-grams. The finding, discussed in Chapter 2, thatbeing in a village with a VO had a greater effecton incomes than being a VO member is relevantto this poverty-alleviation efficiency issue.

In addition to evaluating the efficiency of thepilot poverty program itself—much of which, asnoted earlier, helps the destitute—it would beimportant to monitor the distribution of creditby poverty quintile to assess the share reachingthe poor. This may indicate opportunities foradjustment in the microfinance program. Ideallythis should include monitoring the povertyimpact of internal lending activities.

C o m m u n i t y O r g a n i z a t i o n s

2 3

2 5

Programs

While community organizations are the main vehicle of development,the substance of support for sectoral growth is in four mainprogram areas: microfinance, productive physical infrastructure,

natural resource management, and marketing and enterprise development,together with the human resource development associated with these programs.

The evaluation team assessed each of the fourprogram areas separately, again applying thestandard OED evaluation criteria. This sectionsummarizes the findings of that assessment, thedetails of which are in Annex L.

Microfinance The microfinance component of the AKRSP’swork is about to undergo a major change: theactivities of the Microfinance Section will soonbe taken over by a new national microfinancebank. The AKRSP will be a 60 percent share-holder in the bank (with the AKFED at 20 per-cent, and the International Finance Corporation[IFC] at 20 percent), which would operate bothin the NAC and elsewhere in Pakistan. Theevaluation mission did not appraise the newbank and limited its work to reviewing pastmicrofinance performance and to offering find-ings that are relevant to the operation of the newbank and to the AKRSP’s relationship with it.

The objectives of the Microfinance Program,as presented to the mission, were to:

• Create a large capital base through regularsavings.

• Provide sustainable access to microfinanceservices.

• Devise simple and appropriate financial manage-ment systems for community organizations.

• Improve financial and managerial skills ofthe office staff of community organizations.

• Establish a sustainable financial institution.

These objectives are consistent with the over-all AKRSP mission statement, except that they donot directly cover the equity issue. It is impor-tant that these objectives—but including themissing equity element—are carried over to thenational microfinance bank.

The Microfinance Program Has Been Rele-vant, but Recent Significant Changes Will AffectStrategy. Since the last evaluation, importantchanges have occurred. Since 1997 annual loandisbursal has declined; lending within com-munities has increased; overdues have in-creased; and, in 2000, total savings declined for

44

the first time. These changes, together withthe proposal for the new bank, require thatobjectives and target clientele be clarified, creditproducts be reviewed, and processes of coor-dination between the AKRSP and the new bankbe established.

Efficacy of the Microfinance Program HasBeen Substantial. The program has been animportant complementary element in the devel-opment of VOs and WOs. From the outset, theAKRSP has stressed savings by community organ-izations as an investment in social organizationand as the capital base for investments in pro-ductive opportunities. The processes establishedhave been very effective in reducing transactioncosts and mobilizing savings of poor house-holds. While saving to a certain level was com-pulsory, the microfinance program was helpedfor some time by the differential interest ratesand potential for arbitrage—borrowing at lowrates, saving at high rates—that existed.

Savings and Credits. Recent changes in the sav-ings and credit environment need to be addressedin the new strategy. Total savings declined in 2000to Rs. 429 million from Rs. 431 million in 1999,although it was still rising for the newer WOs.The decline largely has been due to reduced vol-umes of compulsory savings as savings reserveshave built up and to competitive opportunitieselsewhere. However, by June 2001 savings hadrisen again to Rs. 432 million.

On the credit side, while the AKRSP’s creditdisbursements grew rapidly from Rs. 1 million in1983 to Rs. 91 million in 1994, and to a high ofRs. 277 million in 1997, disbursements havedeclined since then to Rs. 224 million in 1998 (Rs.253 million in outstanding loans), Rs. 163 millionin 1999 (Rs. 201 million in outstanding loans),down to Rs. 127 million in 2000 (Rs. 142 millionin outstanding loans). While increased interestrates on all credit products and a downturn inthe national economy have contributed to thereduction in demand, it is mostly a product ofincreased internal lending within communities.

Internal Lending. Internal lending, earliercalled village banking, is self-managed credit inwhich the community organization lends its sav-ings to its own members. Lending terms and con-ditions are set internally by the community.

When internal loans are not repaid, the com-munity organization is faced with a difficultchoice, either deducting the payment from theborrower’s savings, which some organizations arereluctant to do for fear of decapitalizing thecommunity organization, or loan rescheduling,which many do, but which risks decline in lend-ing discipline and erosion of savers’ confidence.Both outcomes risk disintegration of the socialcapital that has been fostered over many years.

It is estimated that in the Gilgit region about18 percent of the community organizations areundertaking internal lending, and in Hunza,close to 50 percent. While this is in keeping withthe AKRSP objective of making villagers self-sufficient, there are doubts about whether itwill be sustainable at present village skill levels.While the AKRSP has limited leverage in theinternal decisionmaking of the organizations, ithas been promoting a set of uniform internallending guidelines and providing training.1 Thiswill need to be continued under the new bank.

Overdues and Write-Offs. Recovery performancehas recently been under threat owing to a highratio of overdues to loans outstanding—above 20percent, with a high of 25 percent in 1999—anda loan loss ratio increase to 4.1 percent in 2000from 1.7 percent in each of the two previousyears.2 Portfolio at risk3 peaked at 7.2 percent in1999 but in 2000 came down to 5.6 percent.

There is a marked regional difference in thepattern of repayments: in 2000, Baltistan Districthad the highest ratio of overdues at 48 percent,Chitral the lowest at 16 percent. The increasedoverdue ratio is largely a legacy of a recent creditoperation failure, mainly involving individualloans (previously called enterprise loans) forpotato production in Baltistan, an enterprise thatfaced production and marketing problems. Thisis a consequence of an attempt in the mid-1990sto rapidly expand the AKRSP’s credit portfoliothrough experimentation with lending products,but without proper appraisal procedures.

The AKRSP has been proactive in reevaluat-ing its loan products and quick to terminatepoorly performing products, however, and ithas now amended its internal policies and pro-cedures. Considering the regional differences inthe overdue pattern, an adequately differentiated

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

2 6

approach, with particular attention to theBaltistan District, is required and appears to bewell in hand. Its effect on repayment perform-ance needs to be closely monitored.

Overall, Efficiency Is Substantial Comparedwith Pakistan and Global Standards. While theoperating cost4 ratio over average portfolio out-standing increased to 11.3 percent in 2000 from7.1 percent in 1998, this is well below theindustry standard of 13 to 21 percent. Theincrease is a result of staff growth and improve-ment of the financial management system. TheAKRSP’s microfinance operations are not directlycomparable with most microfinance NGOs inPakistan because of the small size of their loanportfolios compared to the AKRSP, differentimplementation methodologies, and the limitedtime in operation.5 However, a comparison of theAKRSP with other NGOs in the Pakistan Micro-finance Group6 indicates relative efficiency. Tak-ing a global view, the efficiency of the AKRSP’sMicrofinance Program is better than the indus-try average of 114 microfinance institutions.

Institutional Development Impact of the Micro-finance Program Has Been Substantial. TheAKRSP has established an efficient microfinancesystem and substantially enhanced capacity at alllevels through training. To deal with the emerg-ing overdues issues and to prepare for the tran-sition to the new bank, the AKRSP hasstrengthened its Microfinance Section. As a result,the operating costs of the Microfinance Pro-gram increased from Rs. 2.8 million in 1995 toRs. 19.41 million in 2000. The changes over thepast three years, aimed at portfolio qualityimprovement, have led to an increase in cost perunit of money lent and an increase in the cost

per borrower, while the average amount lent perborrower has gone down and the number ofloans and the credit amount being overseen byeach credit officer have decreased appreciably.

The program has been instrumental in devel-oping local capacity in financial management.More than 3,700 people have been trained inbookkeeping and accounts, including manycommunity organization managers. However,the new bank will present a new skills challenge,coming at a time when AKRSP microfinance isemerging from a period of some fragility.

Subsidy Is Modest or Zero Depending onAssumptions. With respect to financial sustain-ability, the AKRSP’s own Subsidy DependenceIndex (SDI) calculation shows a very favorablenegative 130 percent (Annex G), indicating nosubsidy. However, three coefficients used in thecalculation bias it toward financial sustainability,depending on the interpretation of what theopportunity cost of funds has been, how to treatshared costs within the AKRSP, and how to treatearnings from investment of unused funds (dis-cussed more fully in Annex L). Reworking theindex calculation with alternative assumptions(see Annex G, right-hand column) shows anSDI for the year 2000 with about a zero subsidy.Less favorable alternative methodologies still donot show substantial subsidies. The new bankprojections expect some cross-subsidy from otheroperations to the NAC for about 5 years.

The Future for MicrofinanceClarify the Strategy and Prevent Mission Drift.Both the new bank and the AKRSP need a NACmicrofinance strategy. The evaluation teambelieves that to be consistent with the mission

P r o g r a m s

2 7

AKRSP MBB peer group MBB, all Microfinance Microfinance in South Asia participants Group in Pakistan

Program (n = 9) (n = 114) (n = 10)

Operating costs/average loan portfolio 12% 20% 31% 15%

Operating costs/borrowers $16 $19 $150 $16

Borrowers/staff 389 229 111 N/ASource: AKRSP 2000h.

E f f i c i e n c y C o m p a r i s o n sT a b l e 4 . 1

statement of the AKRSP, the microfinance pro-gram objective for both the AKRSP and the newbank should: (i) maintain a focus on the poor,but not necessarily the poorest of the poor; (ii)maintain a focus on the NAC as the AKRSP clientarea; (iii) provide high-quality financial servicesto the identified clientele; and (iv) maximize thecomplementarities between the AKDN family ofprograms and the new bank. The establishmentof a new bank should neither become the micro-finance objective itself nor absolve the AKRSPfrom setting out a clear strategy for its micro-finance support for the NAC and continuing itscentral role in microfinance execution.

Divorcing microfinance activities from theother AKRSP program components, as the newbank will do, offers some benefits but alsocarries risks that need to be managed. In thepast, the ability to integrate credit activities withtechnical assistance from other program com-ponents was thought to be an advantage. Whileunder the particular circumstances of theAKRSP’s well-coordinated program this mayhave been the case, global experience hasshown that such a tie can actually hurt creditmore than help it, providing the opportunity forborrowers to default on the grounds of faultytechnical advice. With separate institutions—theAKRSP and the proposed new bank—the advan-tage of securing loan repayments as an incen-tive for further AKRSP investments at the villagelevel could be weakened, but the intentionappears to be to maintain close program linksand to withhold investments from defaulting vil-lages. In making projections it will be importantto reflect realistic trends of internal lending andoverdues. Getting these projections right, andkeeping them realistic on the basis of per-formance, will be critical.

Clarify Respective Roles, Especially in InternalLending. It will also be important to clarifyAKRSP and bank roles with respect to guidancefor communities on internal lending, perhapseven regaining some control over the situation.The new bank will need to address this withAKRSP help. Experience suggests that problemswith internal lending will readily spill over intoexternal accounts behavior. Delinquency prob-lems identified early in internal accounts can sig-

nal problems with external accounts while theycan still be fixed.

Conduct Marketing Studies. Mission discus-sions at village level found some dissatisfactionwith loan products, suggesting that further mar-ket research is warranted to contribute to thedesign of new bank products, particularlyregarding the credit needs of women. This is anopportunity to enter a new phase of productdiversification and to move away from the ear-lier, more basic products, or at least to fine-tune existing products. At the end of 2000, andthen again in October 2001, two new productshave been introduced, so there is already evi-dence of increased diversification.

Productive Physical Infrastructure The benefits and impact of the 2,000 small-scaleinfrastructure schemes completed so far (seeAnnex J, tables J.6 to J.9) are substantial andlargely sustainable (about 50 percent irrigation;25 percent roads, bridges, and other communi-cations infrastructure; and about 15 percentmicrohydel). Ingenious irrigation supply chan-nels, often over long, precarious routes acrossmountain slopes, have aided the developmentof 48,000 hectares of new land, about a one-thirdincrease in cultivable land area, and intensifiedexisting cropping. The hundreds of miles oflink roads and 60 bridges constructed by villageorganizations have increased the impact of theexpanding government trunk road system.Microhydels are making a major difference toquality of life, especially in Chitral. The great vari-ety of other schemes chosen by villages indicatesopportunities for more diversified infrastructureinvestments in the future.

PPIs Have Been Relevant. The objectives of thePPI program were highly relevant and consistentwith the overall mission statement. They were:to provide the essential catalyst and social build-ing block for village organization and to respondwith technical assistance and grants to theexpressed needs of the villagers to improve theirlives by investment in physical works. Generally,investments have been equitably distributed.Government has demonstrated its appreciationof their relevance by providing funds for extraschemes to be overseen by the AKRSP.

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

2 8

Efficacy Has Been Substantial. The PPI pro-gram has continued to complete an average ofabout 120 schemes a year, and with the additionof 20 contract engineers, 154 schemes were com-pleted in 2000. (But relevant to the question ofincentives for sustaining community organizationsis that, in 1987, after 5 years of the program, about20 percent of the organizations had a PPI beinginitiated that year, whereas in 2000, with morethan three times the number of organizations andsome decline in number of PPIs initiated, it wasabout 5 percent.) Selection of projects by thewhole community in open meetings conductedwith AKRSP facilitators generally ensures thatthe highest-priority schemes are chosen. It alsoensures collective commitment to provide a sub-stantial share of the costs and labor. Only 4 per-cent of schemes have been abandoned as failures,most of them in Chitral and Baltistan.

However, the rate of completion of the landdevelopment elements of irrigation projects tomake early use of available irrigation water iscause for concern. Reclassifying irrigationschemes as “land development schemes” raisesthe profile of the economically important landdevelopment phase, but it is not clear that thishas yet translated into faster development rates.

Apart from some government construction ofsecondary roads, AKRSP roads and bridges haveprovided the bulk of the tertiary road system.These tertiary roads have greatly increased theimpact of the Karakoram and Skardu highways.

The benefits of microhydels are numerous andfar-reaching, but limited mainly to Chitral, whichhas the most schemes and the highest microhydelconcentration in the world. Initially the main useof the schemes was for lighting to replacekerosene lamps and wood fires, and to powersimple home appliances, but the newer schemesare creating opportunities for small enterprisessuch as stone polishing and agricultural pro-cessing, and for public facilities. There is scopein Chitral for many more such schemes.7 AKRSPmicrohydel staff have provided assistance to avariety of other schemes, even in other countries,projecting the AKRSP approach well beyondthe program.8

Economic Analysis Suggests the Infrastruc-ture Program Is Efficient. The economic and

social impact of the infrastructure program hasbeen substantial: ERRs average 19 percent, basedon AKRSP’s worst-case economic analysis sce-narios (after the omission of one high ERR casestudy outlier in each PPI type). The benefits wereconfirmed in village discussions, which oftenstressed the health benefits and where estimatesof land value changes collected by the missionwere consistent with the capitalization of highprivate internal rates of return into land price.Costs are substantially lower than for traditionalpublic works.

Evidence is mixed on the allocative efficiencyof investments. Average ERRs by type fall in arelatively narrow range, suggesting reasonablysound overall allocation by PPI type. However,the range of ERRs within a PPI type is wider, sug-gesting there may be room for stricter selectiv-ity within categories.9

Sustainability Is Likely, but Microhydels HaveSome Potential Problems. The sustainability ofinfrastructure schemes is assessed as highly likely,but with reservations for some types of schemes.Village ownership of schemes was found to beexcellent. Maintenance of infrastructure is of agenerally high standard: the AKRSP reports that92 percent of the infrastructure projects are effec-tively maintained, with a high figure of 98 per-cent for those in the Gilgit region.10 A recent study(Khwaja 2000) that looked at the determinantsof collective success in the maintenance of AKRSPinfrastructure projects found that adversecommunity-specific factors did affect standard ofmaintenance, but that this could be more thancompensated for by better project design. Noserious problems were reported during villagemeetings, and no major maintenance deficiencieswere observed at project sites visited.

However, three sustainability issues are ofsome concern:• Cost recovery in microhydel schemes gener-

ally just covers routine operation and main-tenance, with no reserve for larger repairs orreplacements. The current ad hoc levies forsuch large expenses seem to have worked sofar, but cannot be sustained in the plannedexpansion to larger schemes. The RegionalProgram Office in Chitral is aware of theproblem, and action is being taken to increase

P r o g r a m s

2 9

tariffs (a similar problem is likely for domes-tic water supply).

• In some irrigation schemes there should beincreasing attention to projecting labor con-straint, especially where there are growingoff-farm employment opportunities. In somecases, village land area increases substantially.

• Water use efficiency is becoming increasinglyimportant, as some villages have experienceddecreasing water supply. While this is not yetwidespread, shortages could pose future sus-tainability problems for some irrigationinvestments.

The Future for PPI Demand Should Be Met, but with Greater CostRecovery. The NAC has a large unsatisfieddemand for further investment in high-priorityinfrastructure, but it should be pursued withinthe framework of increased partnership withgovernment, with greater differentiation betweenprivate and public goods, with increased focuson poorer areas, at scales that villages can main-tain, and with less subsidy (a reduced grantelement). The level of subsidy should be dif-ferentiated by community poverty, public andprivate goods categories (less to the latter), andby resource levels. However, the AKRSP shouldbecome more a social facilitator and technicalassistance agency for infrastructure than afinancier.

Differentiate More Between Public and PrivateGoods and Poor and Less-Poor Communities.The AKRSP should differentiate more betweenprivate and public goods, or the public and pri-vate elements of mixed goods, and make thisexplicit in funding decisions, with the grant ele-ment of private goods being reduced in stages.This would mean that land developmentschemes (including their irrigation supply) andmicrohydels would receive reduced capital sub-sidies, but roads and bridges would continue tobe grant-financed at about current levels. But jus-tifications should be reassessed as part of anoverall AKRSP subsidy review. Levels of povertyshould also be considered. The allocation ofinfrastructure investment should be a major toolin rebalancing the overall program more in favorof poorer and relatively under-served areas, but

as before, only in response to village demandand within village capacity.

The schemes taken up under the programshould normally be limited to those that can beimplemented with no greater AKRSP input thanthe traditional village-based schemes to date, andto those that villagers are willing and able tocomplete and maintain with their own resources.Thus, the prime determinant should be not sizebut pragmatism with respect to what villages canconstruct and maintain. Large infrastructureschemes identified by communities, but beyondtheir capabilities, should be implemented in part-nership with government. The AKRSP shouldconsider moving toward awarding PPIs on thebasis of competitive proposals and the commu-nity record of implementation and maintenance,but with a poverty rating criterion also.

Natural Resource ManagementNatural resource management (NRM) activities,encompassing agriculture, livestock, and forestry,have been highly relevant and have had a sig-nificant beneficial impact in the project area overthe period of the program. NRM activities can beexpected to remain relevant, as about 60 percentof household income comes from farming.

The NRM strategy calls for raising the livingstandards and incomes of the people of theNAC to a level comparable with the nationalaverage for Pakistan. It proposes four priorityareas: increasing area under irrigation, wheat11

and maize improvement, animal husbandry, andcommercial agriculture. Three strategic elementsserve as an overlay: increasing production fromexisting farmland by 66 percent; increasing cul-tivated area by 30 percent; and modestly increas-ing income through ancillary enterprises such asbeekeeping, poultry farming, and trophy hunt-ing. Greater agricultural production is expectedto come from increased cropping intensity, newtechnologies, and improved supply of inputs.Forestry is being played down somewhatbecause of the substantial planting achieve-ments and increase in private nurseries initi-ated with AKRSP assistance.

One premise of the strategy—that productionmust increase by 66 percent in line with popu-lation growth (2.55 percent)—warrants rethink-

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

3 0

ing. In an area as challenging as the NAC, nei-ther efficiency nor sustainability calls for regionalself-sufficiency in food production. Income trans-fer and non-farm income are fully acceptable ele-ments in a strategy for sustainable livelihoods inrural areas. Food import may always be needed.

In the absence of a “control” program area itis difficult to prove that the AKRSP has substitutedfor government, but the total absence of gov-ernment extension suggests that the AKRSP hasprobably substituted for government in tech-nology dissemination. In research, with theAKRSP handling adaptive work only, mainlytesting and demonstrating new varieties, theAKRSP contribution has been more comple-mentary, but complementary to a still weak gov-ernment program largely based on researchstations. There is some evidence of recentimprovements at the Karakoram AgriculturalResearch Institute for the Northern Areas(KARINA; Annex I). If this can be sustained, per-haps with donor assistance, it may provide a basisfor more productive and sustainable partnershipbetween government and the AKRSP in research.

Efficacy of the NRM Component Has BeenHighly Satisfactory.12 Over the only period forwhich good income data are available, per capitafarm incomes increased 2.7 times, from Rs. 2,647in 1991 to about Rs. 7,046 in 1997 (in real terms,1999 prices). Before and after that period, dataare inadequate, but income growth rates wereprobably lower, partly due to weak nationaleconomic growth. The lowest percentageincreases have been in Astore, where infra-structure is a main constraint, and the highesthave been in Chitral (AKRSP 2000c). Efficacy wasparticularly impressive in forestry.

Improved varieties of wheat and maize, 400metric tons of which have been distributed sinceinception of the program, are acknowledged byvillagers to have contributed to raising overallcereal production substantially. Based on thesocioeconomic surveys, cropping intensitybetween 1991 and 1997 increased by about 15percent. Growth in fruits, vegetables, and live-stock has significantly contributed to increasedfarm income.13 Examples of relevant and suc-cessful technology innovations supported by theAKRSP include the introduction of exotic varieties

of cherries, apples, pears, and apricots; the hugeincrease in seed potato cultivation for down-country markets; and continued attention toincreased forage production (alfalfa, oats, maize).

But Some Areas Need Attention. Notwith-standing the generally good performance, aswith all programs, there are some signs of prob-lems. First, while it is difficult to prove, theevaluation team believes some vision has beenlost in natural resource management. Evidencefor this includes the weak explicit linkagebetween NRM strategy and poverty objectives inproject design, the continued lack of attentionto the above-channel areas as an integral partof the system, and the partial response to ear-lier recommendations of both the OED andJoint Monitoring Mission Reports.

Second, the evaluation team observed somevariability in performance of NRM interventionsacross villages and districts, which suggests theremay be room for management improvements.

Third, the evaluation team had some doubtsabout whether there has been sufficient focuson technologies for the poorer households, andfor those at high altitudes, although there has cer-tainly been some attention in breed improvementand fodder security.14 The NRM program needssomething similar to the targeted poverty pro-gram, with differentiation that would identify theparticular needs of the poorer households in eachidentifiable zonal system.

Fourth, sustainability of gains in cash cropswill be fragile if more attention is not given toaddressing crop and livestock pest and diseaseproblems—for example, disease problems inseed potato; understanding and linking mar-keting and production; and enhancing the capac-ity of the private sector input supply system.

Fifth, in double-crop areas there are somesigns that productivity increases are reaching aplateau. This needs attention through integratednutrient management approaches.

Relative Strengths and Weaknesses. Lookingat AKRSP performance in NRM, the evaluationteam assessed strengths and weaknesses withrespect to five attributes generally consideredimportant for technology development and dis-semination. The findings are summarized in thetable below.

P r o g r a m s

3 1

The Farm Systems Recommendations of thePrevious Evaluation. The evaluation mission’sview is that the farm systems recommendationsof the previous mission should have been imple-mented (see Annex B), although there may beroom for debate about the appropriate intensityand scale.15 With the steady increase in off-farmincomes over time, additional focus on a broaderlivelihoods approach is warranted. Focus onthis latter area should now lie within the remitof the NRM component, since there is no othersuitable institutional home for it. Failure to adoptthe systems approach appears to be explainedby a combination of concerns about costs, lackof a comprehensive NRM strategy, lack of coreleadership in NRM, and lack of strong man-agement commitment to NRM relative to otherpriorities.

Livestock Development in Above-ChannelAreas.16 The AKRSP’s role has been importantthrough the afforestation program and throughfacilitating contacts between communities and

NGOs such as the IUCN and the WorldwideFund for Nature (WWF). But with respect to live-stock productivity, the AKRSP has focused on thebelow-channel area almost exclusively. Yet nearlyall the weight gain of most livestock comesfrom above-channel forage, and livestock are thebiggest components of farm income. More focuson this area, and the interactions with below-channel areas, is warranted. A shift in this direc-tion by the AKRSP would complement theongoing efforts of the IUCN and WWF.

Efficiency Has Been Satisfactory. The eco-nomic analysis of the NRM component alone,excluding the AKRSP overheads, which havebeen applied to the overall economic analysis,suggests a rate of return of 24 percent. Thisquite high rate of return is fairly typical of NRMinterventions with a high level of technology con-tent. The economic analysis also indicates thatrelatively modest shares of aggregate increasesin net farm income attributable to AKRSP, on theorder of 10 percent, would give an adequate ERR.

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

3 2

Attribute Strength Weakness

Integration All disciplines of crops, livestock, and forestry integrated Limited central leadership in NRM to play mentor as well

into NRM; dissemination processes devolved to regional as strategic planner.

program offices.

Flexibility Regional management empowered to make periodic reviews Most NRM staff still target-oriented; little time left for

and change, when needed, resources allocated for specific creative activities or pursuing ideas that would lead to

activities; can be very responsive to the community needs. generating new technologies.

Relevance A number of examples of relevant and successful A few areas/farming systems still do not have appropriate

technologies disseminated by the AKRSP can be found in technical packages and associated support (single-crop

the field. zones/above-channel farming systems; remote communities

with marketing problems).

Partnerships Very strong in building partnerships with communities. Relatively weak in promoting partnership with government

Emerging capacity in building partnerships with private agencies.

sector input suppliers.

Institutional Over the full program, solid support from management; Support for NRM declined somewhat in recent years.

support transparent policies; goal-oriented.

P e r f o r m a n c e i n N R M : S t r e n g t h s a n dW e a k n e s s e s

T a b l e 4 . 2

Environmental Sustainability Is LargelyUnknown. No broad study has been done of theenvironmental costs and benefits contributed bythe people of the NAC to this nationally veryimportant river basin. Such a study is needed toassess the net services provided. With respect toaccelerated soil loss arising from land use prac-tices, sustainability issues do not appear to be amajor concern in comparison with the mass wast-ing of material arising from natural weathering andcollapses of this geologically young mountainregion. While interventions such as afforestationcan temporarily hold more soil, and while indis-criminate tree cutting (arising from ill-considerednew roads, for example) can do the reverse andalso contribute to significant biodiversity loss,the likelihood is low of human activity raising orlowering the total silt load coming out of suchareas by appreciable amounts. Nevertheless, a bet-ter understanding of the broad environmentalrelationships is warranted.

With respect to pesticide use and fertilizerrunoff, there are no data. Training to prevent theindiscriminate use of inputs and further work onintegrated crop management techniques shouldbe pursued as much for profitability as for envi-ronmental reasons. However, the scale of fertil-izer and pesticide use is still quite modest andit is unlikely to present a major problem yet. Butthe issue will need careful watching.

Future Directions for NRMThe following is a very brief summary of pro-posed future component program directionsthat are more fully covered in Annex L:• Collaborate with government to initiate

Research Policy and Commodity ResearchCoordination Committees (less challengingthan zonal committees at this stage) andResearch Extension Liaison Groups in eachzone, with farmer, researcher, and NGO mem-bership—including the AKRSP. Initiate aphased introduction of Participatory and Col-legiate Research. Participatory research rep-resents a stage beyond simple participatoryvariety testing, bringing farmers more into theexperimental process.

• Revisit the farm systems recommendationsand pilot a modified system. In addition,

increase attention to three areas of farmingsystems integration: integrated soil manage-ment, integrated crop management, and inte-grated water management.

• Extend the work in single-crop areas, espe-cially in Astore. Develop it as a special,poverty-focused program using a low exter-nal inputs approach with a focus on improvedvarieties and seed multiplication and supply.

• Do more for the above-channel areas. Whilea decision to do more in these areas may bepremature before most of the findings of theongoing research program are in, this optionshould be given serious consideration as soonas possible.

• In partnership with others, carry out an analy-sis of the environmental costs and benefits ofupstream activities in the river basins, meas-uring impact at the point of exit from the NAC.This may need to be done in stages, but theobjective would be to contribute to an over-all assessment of the net upstream environ-mental services provided by the people of theNAC.

• Give high priority to building the capacity ofsuppliers for provision of quality inputs andtechnical advice. Increasingly, suppliers shouldbecome an element in a more pluralisticextension service, gradually moving beyondtheir present limited roles toward becomingmore substantial and stable enterprises.

Marketing and Enterprise Development Marketing and Enterprise Development Is Rele-vant, but Objectives Need Clarification. Market-ing and enterprise development is clearlyrelevant for a remote region in a diversifyingeconomy. However, the objectives of this pro-gram component need to be clarified and linkedmore directly to the overall mission statement.Activities have ranged so widely that theapproach appears to have been ad hoc, ratherthan guided by some clear principles. There isa need to maximize the AKRSP’s leverage, butto be more efficient in picking the winners.

Achievements Have Been Modest. The origi-nal marketing component concentrated on reduc-ing losses of perishable products—fruit (freshand dried), livestock, and grains—and on input

P r o g r a m s

3 3

stores. Farmer training and credit were pro-vided, but much of the activity was undertakenby the AKRSP and directly primed with grantfunding. Early results showed few successesand some spectacular failures, especially wherethe AKRSP took on an agency function in mar-keting perishable produce.

From the mid-1990s the emphasis swungtoward enterprise development, with the rangeof activities broadening away from the AKRSP’srural roots. Given the diversifying economy,this was a relevant shift for the NAC, but it hastaken activities into areas of lower AKRSP com-parative advantage. Notable successes to dateinclude apricot drying and packaging, whichhas greatly added to producer returns for thisimportant crop; seed potatoes, which startedwell, but now has run into difficulties; vegetableseed production, which has grown to quite alarge enterprise with contract growers, but is stilla “project” striving for profitability and privateownership; the shu fabric/clothing enterprise,which also needs to become private; and agri-cultural input supply shops, which are alreadyin private ownership.

The latest strategy, which the evaluation teamsupports, focuses on providing additional busi-ness development services for small- andmedium-scale enterprises, mainly through part-nership with existing business service agenciesand the private sector. However, the limitedefficacy of the program to date does raise thequestion of whether, for enterprise and marketingdevelopment, roads investment is not the mostreliable bet. Value-added impact comparisonwould be instructive.

Efficiency Has Probably Been Modest. Themission did not have the resources to evaluatethe efficiency of such a diverse collection of activ-ities. But with many starts and limited successes,efficiency has almost certainly been modest. Itis to be expected that this component wouldneed a start-up subsidy as the AKRSP’s traditionalstrategy emphasizes experimentation, trial anderror, starting small, admitting failures, and mov-ing on if needed. The marketing and enterprisedevelopment component has epitomized thisapproach. The new partnership approach shouldprovide a more direct route to establishing viable

enterprises, with partners being selective in theirchoice of enterprises.

Institutional Development Impact Has BeenModest. The AKRSP faces its greatest enterprisedevelopment challenge in devising ways toensure that its activities become institutionallysustainable. The earlier cluster marketing asso-ciations have tended to wither once the AKRSPis no longer the driving force. With skilled staffand ample initial resources, much can beachieved, but this is not development unless itis rapidly institutionalized, preferably as a prof-itable private enterprise or in partnership witha corporate patron. Within the AKDN thereappears to be potential synergy in more closelyallying AKRSP’s enterprise development activi-ties with AKFED resources.

The Way Ahead for Marketing and EnterpriseDevelopment• Evaluate the impact of training. The AKRSP

has provided a wide range of relevant train-ing in enterprise skills such as business man-agement, hotel management, bookkeeping,post-harvest management, broiler farming,fruit and vegetable processing, poultry feedand chick supply, embroidery, carpentry,auto mechanics, motor/generator rewinding,mining, gold panning, food processing, com-puter technology, cobbling, and mineralappraisal. The skills developed are not nec-essarily lost with the failures of individualenterprises. The impact of this training onincome generation needs to be evaluated inorder to prioritize future training activities.

• Stay the course with agricultural marketing.Marketing of agricultural products remains aprimary constraint and should be reinvigo-rated to a level at least equal in priority withmore recent enterprise initiatives. This maycall for a staffing review to ensure the avail-ability of appropriate skills.

• Develop a decision framework. A decisionframework should be developed and keptupdated to identify those sectors, subsectors,and product categories likely to be most com-petitive, as well as add the greatest value tothe economy of the NAC. Greater prioritiza-tion is desirable, with more resources going

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

3 4

to fewer prime areas to avoid being spreadtoo thinly. The use of a social accountingmatrix is worth exploring.17

• Tourism should be a priority area for AKDNcoordination. Tourism in the region’s dra-matic physical environment has grown inrecent years, with some AKRSP assistance forvillage hotels, but it would be desirable to havea more defined strategy in the sector thatembraces other relevant AKDN agencies. Apartfrom traditional trekking and climbing tourism,the region’s natural wonders and biodiversity

suggest there is substantial potential for eco-tourism if related infrastructure can be pro-vided and security can be maintained.

• Look outward at migration as part of the NACstrategy. Focus increasingly on marketableskills for migration outside the NAC within abroader livelihoods strategy. As has long beenthe case, the future for a significant propor-tion of the population does not lie within theNAC. Support and assistance in employmentinformation and in the migration processitself may be warranted.18

P r o g r a m s

3 5

3 7

Future Directions

The Mission Is Valid, the Direction Needs to Change

The AKRSP is engaged in an intensive dialogue on strategy both inter-nally and with donors in preparation for the next phase of funding.A main purpose of this evaluation is to contribute to that dialogue.

This stage in the AKRSP’s evolution represents a significant watershed,coming at a time of increased donor fatigue and continued concerns aboutefficiency, impact, and future strategies; potentially far-reaching changes inthe role of community organizations, including the establishment of DehiCouncils; and the creation of a new bank. After nearly 20 years of successful

work that has contributed greatly to the social andeconomic growth of the NAC, the AKRSP is ask-ing itself, and being asked by its principal part-ners, to plot a course for the years ahead that isnot simply a linear extension of the road just trav-eled. Although there has been great change forthe better in the NAC there is still far to go, withmost indicators of development still lagging thosein Pakistan as a whole. Based on 1998/99 data,the average person in the Northern Areas is stillliving on about 50 cents US per day. Thus, thereseems little reason to change the AKRSP’s basicmission, most recently stated as “to promoteequitable and sustainable improvement in thequality of life of the people of the NorthernAreas and Chitral” (AKRSP 2000a, p. 3). But,given changes in the development and donor

environment, there is a need to adjust the ap-proach to achieving that mission.1

The route to sustainable growth of the NACeconomy lies mainly through private sector de-velopment, an area where neither the AKRSP norgovernment has yet found a way to be effective.Both the AKRSP and government have roles toplay; yet both need to learn how to foster theprivate sector without handing out extensivesubsidies or encouraging investments that lackthe economic prerequisites for success. As apart of this, out-migration and seasonal em-ployment migration are long traditions. Thesebehaviors are not only to be expected, but alsoto be encouraged and assisted, as part of theprocess of managing the employment problemwith a national perspective.

55

The foregoing points to the need for gov-ernment capacity to be strengthened.2 The aimshould be to build public sector development ca-pacity with the support of AKDN institutions.There is room for both, but the balance shouldchange (as indicated in figure 5.1). The institu-tions of the AKDN and government have a jointinterest in expanding public sector capacity3—expanding it to the point where it can respondeffectively to the increasingly articulate demandsof local communities, now widely organizedinto effective village-level bodies and the newDehi Councils.4 In such a partnership, the AKRSP,because of its superior organizational skills,strong community-based record, and accumu-lated social capital, should lead the way.5

Form a Development CoalitionThe change of direction would call for the fol-lowing three strategic thrusts in response tothree challenges:6

1. Agree and implement a common, coordi-nated approach to development in the NACwith the other organizations in the AKDN.7

This would call for attention by donors to theincentives for coordinated action.

2. Take the lead within the AKDN in agreeingand implementing a clearly defined, mutuallydependent development coalition foundedon a partnership with government and the pri-vate sector. The intention is to raise the de-

velopment capacity of the public sector withina set period, while simultaneously increasingthe role for the corporate sector.8

3. Deliver both of the above while: (a) main-taining and enhancing the strength and vital-ity of an increasingly diverse range ofcommunity organizations that still lie at the coreof future development, whether supported bythe AKDN or the government, and (b) devel-oping its own policy research and think tankrole. Within this thrust there should be an el-ement of “back to basics”—reemphasizing theoriginal concept of empowering community or-ganizations to improve the welfare of theirmembers through their own efforts. This is athrust that has been under some threat overthe years from supply-led donor support.

This would leave the NAC more or less equallyin the hands of the community institutions andthe people they represent, the private and cor-porate sector (of which AKDN institutions are apart), and government—each enabled to pursuethe development challenge in accord with its owncomparative advantage. It would leave the AKRSPand the rest of the AKDN with a quantitativelysmaller, but still pivotal, position in the NAC. Theprocess itself would need to be initiated by acoalition workshop of all parties, with two orthree readily achievable and measurable actionsagreed in each area of operation. Ideally, the stepstoward increased coordination and consolidationwithin the AKDN should be taken before thecoalition is fully established. But taking these stepsshould not delay action on the coalition.

In parallel, the AKDN family, including theAKRSP, should increasingly be looking outwardfrom the NAC, toward broader partnershipsaimed at non-farm income sources—for example,facilitating rural/urban business partnerships,linking with small urban goods manufacturers,increasing artisanal training, getting national-level as well as NAC-level employment oppor-tunity information into villages, and facilitatingurban links, perhaps through city transit hostels.

As with all challenges, there are risks. In mis-sion discussions, doubts were raised about thecapacity of government to perform effectively asa partner. The concern is valid, but it is more rel-

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

3 8

F u t u r e S h a r e o fD e v e l o p m e n tC a p a c i t y( i n d i c a t i v e )

F i g u r e 5 . 1

Current In seven years In fifteen years

Corporate sector

Civil society

AKRSP/AKDN

Government

evant to the rate of change than to its direction.9

Adequate coordination within the AKDN mayremain elusive, and efficiencies may not be cap-tured. Government may be inconsistent or weakin its commitment.10 National economic per-formance may be slow to improve, and the cli-mate for private sector development or policyreforms may remain difficult. Such institutionalfailures are one of the challenges of develop-ment, but they do not constitute a case for notmaking an energetic start.

The International Year of Mountains in 2002represents an opportunity to connect redirectionand learning to broader global activities in geo-graphically similar areas. To minimize risk,11

plans and activities must be incremental, and dis-cussed and agreed beforehand, and objectivesmust be kept to modest, actionable, and meas-urable steps. A pragmatic strategy is neededwith a clear direction, an initially modest ex-pectation, clearly defined monitorable steps ina limited number of program areas, and a focuson capacity building and devolution. The roleof the board will be important.

Regular monitoring is particularly importantwhere there are risks and uncertainties. The goalof building mutual trust and confidence amongall the partners suggests that a joint monitoringunit should be created to measure and report onthe coalition’s progress. The government has noeffective monitoring capacity; it would be pru-dent, therefore, to help strengthen government’scapacity. For the most part, success would bejudged by regular assessments by villagers them-selves of their satisfaction with services providedby the principal agents in the coalition.

New skills in AKRSP and AKDN institutionsmay be needed. Goals and commitment willneed frequent reinforcement from senior man-agers and governing bodies. There will also beneed for crucial political and advocacy skills tobe deployed by all sides, and especially by theAga Khan Foundation.12

The donors should consider continuing aidto the AKRSP and other members of the AKDNfor a further defined period. This support shouldhave the following aims:• Assist AKRSP efforts to help the NAC achieve

some threshold of poverty achievement such

as the Millennium Development Goals orPakistan poverty threshold.

• Establish a separate fund to be drawn downby government as it seeks to fulfill its shareof the bargain.13 Without the application ofexternal resources it is unlikely that the nec-essary changes in government commitment,behavior, and programs will be secured withina reasonable time.

• Support a major push on the lagging Astorearea focused on infrastructure development.

• Where possible, contribute to the proposedendowment fund. But donors will rightly beunwilling to enter into an open-ended com-mitment. Hence, clear milestones would needto be jointly articulated.

Shaping the StrategyThe strategy should describe the Pakistan andNAC development context and outline a vision ofwhat the AKRSP might be expected to look likein 10 to 20 years. Probably the AKRSP would havethe following characteristics: it would be smaller,continue to assist communities to improve theirown welfare through their own efforts, recovermore of its costs from fees, focus even more oncapacity development, be as much a creative thinktank for new development ideas for NAC andPakistan as it is an area development program, ven-ture beyond Pakistan to seed social skills in othercountries, develop creative policy options, be aplayer in a larger NAC team, and raise a significantshare of its funding from consulting services. Withits clients it would be more a “you call us,” demand-led service organization (relying, in due course, onvastly improved communications) than a supply-led organization. Above all, the AKRSP would besignificantly less indispensable to the NAC, or in-dispensable in an entirely different way. AKRSP sizewould be an outcome of strategy, local demand,and changing needs, and not an end in itself.Donor strategies may or may not coincide with theAKRSP strategy in the longer term. The objectivesand the field of play of the two are different.

The strategy should emphasize two key de-sign principles: efficacy and efficiency. Efficacyshould increasingly have two dimensions—theefficacy of the AKRSP itself and the efficacy ofthe partnership as a whole. There would need

F u t u r e D i r e c t i o n s

3 9

to be clear definition of the primary and subsi-diary program and objectives, explicit statementsabout their relative importance, measurable in-dicators for each, and procedures for carryingout the measurement.

Efficiency would be pursued largely throughdifferentiation according to individual commu-nity needs and comparative advantage withinthe AKDN family and within the partnership.Criteria agreeable to all parties would bedeveloped that would allow effective differen-tiation of communities and would also addressthe problem that the most mature village or-ganizations tend to be those with majority Is-maili populations and that, in Chitral especially,the left-out non-Ismaili villages are among thepoorest. The support to be provided to DehiCouncils would be differentiated in step withtheir evolving role. This is a natural area ofpartnership with government.

Potential vehicles for the partnership elementwith government are the new generation of so-cial services projects, such as the Poverty Alle-viation Project, which channel public fundingthrough NGOs. (The AKRSP is playing a deliv-ery role in this project.) But government anddonors need to provide support within theseprojects for government capacity building sothat the future is not an unequal partnership.

Developing a Strategy for Enhancingand Sharing AKRSP LearningThe AKRSP should not rest on its laurels butshould develop a deliberate strategy to broadenand share its learning more widely. Currently, thesharing of accumulated knowledge appears to besomewhat opportunistic, depending mostly onAKRSP staff moving on to other jobs (see AnnexA) or on hosting interested parties from otheragencies—although the internship program doesrepresent an important element of a strategy.Sharing learning cannot simply be supply-driven.

The AKRSP should determine a target audienceand develop a deliberate program to disseminateits knowledge, including areas in Central Asia out-side Pakistan where similar geographical andsocial circumstances prevail. Expanding the in-ternship arrangement toward exchange intern-ships to other organizations worldwide would be

one approach with potential for a two-way flowof knowledge, which would also benefit theAKRSP through infusion.

The AKRSP can learn much from other prac-titioners, not least those who have adopted andmodified the AKRSP approach. Such occasional“ground truthing” of its own practices in useelsewhere could be mutually beneficial (andcould help to divert suggestions of AKRSP arro-gance that are sometimes heard from observers).The 1999 Joint Review Mission noted that al-though external perceptions of the AKRSP areconsistently positive and the development modelis acknowledged as important, yet (still) theAKRSP should be open to learning from other de-velopment initiatives.

Organizing for PartnershipThe AKRSP has been an institutional successpartly because of wise leadership that, for 20years, has emphasized close links to the field,where its principal clients live, and fostered astrong sense of internal organizational equity.These same values will serve the AKRSP well asit seeks to adjust its organizational structure tonew realities. Current proposals include changesto the structure of field offices and the additionof a Policy Unit. This proposed unit has a wideremit and it would be prudent to rank its activ-ities and initiate them progressively as staff withthe requisite skills are identified.

The AKRSP is right to take the opportunity ofa change in direction and strategic focus to ad-just its organizational structure. In reorganizing,the AKRSP should continue to be guided by or-ganizational equity, but also by the demands ofthe overarching goal of forging a developmentcoalition in the NAC through interlocking part-nerships with all the key players. Reorganizingfor partnership will call for the closest possibleinteraction with the key institutions in the NAC,especially government and the AKDN. This goalshould be given the greatest weight in framingthe future organization of the AKRSP.

The proposed changes in the AKRSP’s fieldstaff management and greater devolution withinthe NAC thus seem appropriate, while dividingthe Core Office between Gilgit and Islamabadseems inevitable given the differential pull of local

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

4 0

and wider objectives. However, giving greatestweight to partnerships and considering the mixof work during the next few years strongly sug-gests that most of the staff in the Core Office (in-cluding the Policy Unit, if created) should bebased in Gilgit. In this respect, circumstances arechanging rapidly. With the help of the Internet,activities that were once “location bound” arenow “footloose.” This locational freedom allowsactivities and staff to be sited where their com-parative advantage is greatest. Using these prin-ciples to guide organizational change will helpensure the AKRSP’s continued relevance to thedevelopment challenges of the NAC.

Social Accountability and GovernanceGlobally, the trend in rural development hasbeen toward increased accountability and ben-eficiary participation to ensure that local needsare adequately reflected in program response.The new strategy being developed offers anopportunity for the AKRSP to shift toward amore open and transparent mechanism to im-prove the responsiveness of the program. Thiscould be done at many levels—from, at thehighest level, increased NAC representation onthe board; to increased village representationalpresence at program management or componentmanagement meetings; to solicitation throughsurveys of individual views, which are likely todiffer by gender, household, and income groupat the village level. This could be a step towardsome degree of membership-based spin-offsinto new specialized organizations.

Future Strategy—CommunityOrganizationsThe proposed shift toward a “functions-based ap-proach”—in which the choice of institutionwould be based on suitability for the functionsproposed14—is an appropriate evolution, pro-vided the needs of women and the poor are ad-equately addressed. There is some risk in afunctions-based approach that such considera-tions will be sidelined. The AKRSP needs to de-velop a vision and strategy for communityorganizations that will bring women into themainstream of village decisionmaking and de-termine the balance between support for or-

ganizations and support for enhancing individ-ual competencies. It needs to pursue efficiencythrough differentiation of village and organiza-tion support and through selective withdrawal,and to better connect women’s developmentwith other AKRSP and AKDN services. And itneeds to address the absence of core leadershipin the women’s program.

The Need for Incentives for ContinuedCommunity ParticipationContinued community participation on the scaleseen in the past will be unlikely unless incen-tives for participation can be maintained. Thiswill call for even greater sensitivity and open-ness to community demand and demand of dif-ferent subgroups, especially the poor. The simpleenvironment of the past is changing. The initialflush of PPIs is past, microfinance is shifting toa new bank, and new community organizationstructures are evolving. Support and training inagriculture, livestock, and forestry will remain im-portant incentives, but even here there is likelyto be some trend toward individual rather thancommunity motivation.

Nevertheless, there will be a number ofemerging areas where the AKRSP can enhanceincentives for community participation for thebenefit of the poor and communities as a whole.These areas would include increased attentionto women where AKRSP interventions are at anearlier evolutionary stage; increased facilitationof public funding, particularly in larger infra-structure; support for emerging organizationswith new functions and different incentives, in-cluding Dehi Councils; increased focus on high-value-added commodities; continued selectionby the group of individuals for training in newskills or for exchange visits; increased attentionto skills relevant to migration and non-farm in-come and facilitation of migration to ease ad-justment costs; continued attention to improvingthe efficiency of marketing chains; increasedfacilitation for other AKDN agencies to meet abroader array of community demands; supportfor federations to add incentives related to scale;child care for women going to village meet-ings; broader self-improvement training and ac-tivities; some of the increase in cost recovery

F u t u r e D i r e c t i o n s

4 1

placed in a community pool to offer as match-ing funds; and increased support for links to theprivate sector—for example, for bulk input sup-ply and marketing. In the new, more fluid en-vironment, the analytical starting point for newinitiatives needs to be community and householdincentives that accommodate gender differences.

Differentiation and Withdrawal There has been surprisingly little new analyticalexploration of how to either revitalize support forvillage institutions or to withdraw from them ordifferentiate. On one hand, there seems to be im-patience to get away from intensive village inter-actions and into work that is more urban, regional,and national, but there is also a feeling of re-sponsibility to organizations that are not yet readyto stand on their own. A strategic question is howto balance the desire for creative institutional di-rections while not abandoning the still-needed sup-port for the original community organizations.The answer, for the present, seems to be to allowdifferent parts of the organization to provide sup-port in different ways, with FMU staff continuingtheir work in the villages, leaving the regional andcentral policy departments to address new themesand challenges. This ought to be set more clearlywithin a strategic framework aimed at differenti-ation of support based on needs,15 with a 10- to15-year vision shared by all players within theAKRSP and its associated partnerships.

Most well-established development programshave village withdrawal strategies. These arebased on the notion that villages reach a pointin their development where they can inde-pendently access resources (inputs, financialservices, technical advice) through links withgovernment or other institutions. Phased with-drawal from such villages allows agency re-sources to be redeployed to less-developedvillages. The AKRSP still maintains services towell-established community organizations that in-dicated to the team that they could manage with-out further input, or at least with limited support.

There seem to be two difficulties for the AKRSPwith withdrawal at the village level. One is thatdifferentiation on the basis of need (gradual with-drawal from the villages most capable of ac-cessing outside support for their own welfare)

would tend to graduate first those villages withmajority Ismaili populations.16 The other is thatthe AKRSP has now diversified into enterprise ad-vice that is inevitably aimed at an entrepreneur-ial class in better-off rural areas and in towns. Theissue is whether it is possible to combine equityobjectives, which would imply withdrawal frombetter-off villages in order to redeploy resourcesto poorer ones, with efficiency objectives, whichwould imply staying in many villages and offer-ing enterprise or more generalized poverty sup-port services according to need. Currently, theAKRSP seems to favor the latter strategy. In onesense, equity objectives are taking second placeto efficiency objectives. Withdrawal is complicatedby the uneven development of village andwomen’s organizations. Even if the AKRSP agreedthat it would no longer visit many organiza-tions—but would continue a “you call us” serv-ice—this would not solve the problems for manywomen’s organizations, which may be in thesame village. Few women can easily visit theFMU to request services. But while withdrawalfrom villages may be difficult, there should be awithdrawal strategy for groups, regardless of theoverall status of a particular village.17

Functions-Based and Supra OrganizationsMake Sense, but Gender Concerns NeedAttentionWith its budgetary pressures, the AKRSP cannotcontinue to support close to 4,000 individualcommunity organizations to the same degree itdid in the past, but it could build up the man-agement capabilities of LDO representatives,who would in turn help village institutions.18 TheAKRSP has concluded, correctly, that this processcannot be pushed too fast (the same was foundin the India program), but that it is an appropriatedirection. While appropriate, however, thereare risks. The AKRSP should focus on ensuringthat women gain rather than lose from such ashift toward LDOs—there is a risk that theywould lose from power shifts to higher levels.This would need close monitoring.

Balance Community with CompetencyA more radical option is for the AKRSP to shiftaway from supporting organizations, toward a

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

4 2

greater focus on the development of individualcompetences—leaving organizations with limitedsupport to grow or die as they will, albeit ben-efiting from any individual skills imparted. Tosome extent the AKRSP is already moving towardsuch a competencies model—for example, in itsenterprise work and somewhat in its proposedshift toward function-based groups. Compe-tency models are knowledge-based, offering in-formation, linkages, and skills. This is differentfrom the organizational capacity developmentmodel. Currently, the AKRSP is doing both—itsupports organizations while also responding toknowledge needs. Has the AKRSP got the bal-ance right? The answer for now is probably“yes,” but the AKRSP should be aware that thereare tradeoffs here. The strategy should ac-knowledge this and map out a position thatmay be different for villages at varying levels ofmaturity.

Influencing DevolutionThere is a major role for the AKRSP in influencingdevolution. The AKRSP is to be commended forresponding quickly to government’s request toassist with the establishment of the Dehi Coun-cils and—not without risk—to support the in-troductory training programs for Dehi councilors.The extent to which support for the Dehi Coun-cils becomes integral to the program will haveto be determined by the rate and direction ofevolution of these councils. Depending on thatdirection, AKRSP support could go further.19 Inpartnership with government, it could monitorprogress and provide feedback as the processevolves. It could do studies on Dehi Council evo-lution and attitudes of villages aimed at influ-encing policy. It could learn from devolutionexperiences in other parts of Pakistan and glob-ally. It could support links between Dehi Coun-cils and the AKRSP’s community organizations,which are already facilitated by many villagersholding positions in both organizations. It couldfurther develop capacity building, including fos-tering training visits of skilled councilors to othervillages. There are risks of excessive political in-fluence from above on what should be a bottom-up process. This is not a reason to remain atarm’s length; it is a reason to be involved.

Defining a Future for Women’s OrganizationsThe women’s program lacks vision, direction, andcore leadership. A key question for the vision andfuture strategy is whether to try to push women’sorganizations to “catch up” with village organiza-tions as independent and autonomous institutionsor to look for an alternative path. One optionwould be for the AKRSP to develop a partnershipwith government in a program of training and sup-port to women Dehi councilors. Although it is un-clear how significant the new Dehi Councils willbe, they present an opportunity that should not bemissed for women to participate with men in de-cisionmaking. If this turns out to be the case, theAKRSP could then gradually shift the emphasisaway from community organizations and focus forthe next phase on women’s skills and abilities asDehi councilors, so that they can become more ef-fective advocates of their own interests.

The AKRSP has been slow to create a broadplatform for women within its own carefully fos-tered organizations. In the Northern Rural Sup-port Project (NRSP) about 10 percent of groupsare now mixed, so it would appear to be realis-tic for the AKRSP to move in this direction also.Furthermore, as noted by Tetlay and Raza (1998),with growing out-migration for employment,households are increasingly becoming “femalemanaged.” Focusing on opportunities and sup-port for village-selected women leaders will un-derpin a new strategic direction for the women’sprogram that should include identifying moreeconomic opportunities for women.20 Some vil-lages already have women who are educated andhave public roles—and in due course Dehi Coun-cil opportunities will increase their number.There is also an increasing number of young menwho support women’s development.

Gender-Related IncentivesAs for the program more broadly, it will be im-portant to seek out incentives for changed be-havior with respect to gender issues—incentivesfor men, including religious leaders, and forwomen. For men, incentives may include morefavorable consideration of PPIs where increasedparticipation by women is being accepted and,for religious leaders, due consideration of theirproposals for development. For women, incen-

F u t u r e D i r e c t i o n s

4 3

tives may include increased credit opportunitiesand water supply for communities where womenare taking a more active role.

The Women’s Program Needs Leadership andConnection with Other AKRSP ServicesSome observers have suggested that there hasbeen some degree of regression on gender overthe past decade. The mission’s view is that therehave been advances in some aspects, but lossesin others. The task is to bring together andmake coherent what is currently separate—thewomen’s and the men’s programs. This taskneeds to be overseen by a senior member of thecore team with expertise in management and ingender programming. Gender expertise cer-tainly exists within the AKRSP’s current staffinggroup, but this post will also require consider-able management and personnel experience.There is some reluctance to recruit from down-country on the basis that experience gainedthere may not be relevant. However, there maybe advantages to having someone from down-country, as they are more likely to have widenational and international experience. The issuesthe AKRSP currently faces are common to mostwell-established development agencies in SouthAsia that started with separate programs formen and women but have since developed co-herent and integrated objectives and structures.

AKRSP organization and staffing needs to beadjusted to help bridge the many gaps women

experience in their attempts to fulfill one of theirmost oft-quoted objectives: to earn more money.There are two important constraints, workloadand organizational coordination. First, thewomen’s social organizer (WSO) has to mediateevery contact between women’s organizationsand FMU specialist departments, as she is usu-ally the only woman staff member. BecauseWSOs often serve more than 150 organizations,they cannot afford the time to visit often. Second,the Enterprise, Learning Support Unit, and Micro-finance Sections seem to function as separate op-erations, not as a coordinated service. Increasednumbers of women staff within these sections,and improved coordination, would help.

Poverty TargetingThe AKRSP’s recently initiated pilot poverty ex-ercise is an appropriate learning strategy to seekto improve poverty targeting and should becontinued for several years. The challenge, how-ever, is to achieve cost effectiveness that couldbe scaled up. This will call for intensive moni-toring and rapid adaptation to pursue the lever-aging of AKRSP skills to the maximum. The useof at least two approaches, one more intensive,one less intensive, would help to answer the ef-ficiency question. The question to be answeredwill be whether the poverty impact of the pilotintervention is more cost effective for poverty al-leviation than the same resources put into themain AKRSP program.

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

4 4

A new vision has a cascading impact. Suppose it was decidedby the AKRSP that the vision was to now help women taketheir place as public citizens and as members of village insti-tutions, not just in women’s organizations. Suitable AKRSP pro-grams and partnerships would then be designed to contributeto this goal. The first task might be an interactive program to in-crease men’s support and understanding for this shift. TheAKRSP might then offer to train women Dehi councilors andboard members of LDOs. It might be agreed with village or-ganizations that, where appropriate, most village-level projectswould be channeled through both men’s and women’s organi-zations in a new partnership. In the interest of improving

women’s economic status, the AKRSP Enterprise DevelopmentSection might develop a special strategy and program to pro-vide relevant advisory and financial products for women. TheNRM program might devise a program to focus on technologiesin which women had a substantial role and income benefit. Thepoint is that a new vision and strategy for the women’s programshould affect the whole AKRSP program, not just a special sep-arate segment for women that is designed and implemented sep-arately from male activities. Thus, a new route map for allAKRSP sections would be needed to ensure that they all con-sider specific activities for women, with the participation ofwomen decisionmakers.

T h e C a s c a d i n g I m p a c t o f a N e w V i s i o nB o x 5 . 1

F u t u r e D i r e c t i o n s

4 5

Pa

ki

st

an

:

Th

e

Ag

a

Kh

an

R

ur

al

S

up

po

rt

P

ro

gr

am

Fi

gu

re

5

.2

4 7

The following is a list of where AKRSP professional staff went to contribute to development throughother organizations:

ANNEX A: AKRSP SKILLS SEEDING

ANNEXES

Destination Number

Globally

AKF outside Pakistan 1

UNDP 2

World Bank 2

CGIAR System 1

Within Pakistan

Consultancy 2

Rural Support Program, including the NRSP 6

UNDP 1

Other NGOs 4

University 1

NAC government 1

Other AKDN institutions 3

IUCN 3

Banks 1

Total 28Note: There are a number of others, but their current work is not known to AKRSP.

Source: AKRSP HR staff.

4 9

Responsiveness to earlier recommendations hasbeen modest.

The Second Interim Evaluation of March 1990noted some weaknesses in the response to therecommendations of the first report, whichincluded:• Integrate infrastructure production and mar-

keting and the need to identify and testmarkets.

• Place more emphasis on households as theunit of observation, including the study oflabor profiles.

• Group farm-level data according to char-acteristics, location, and agro-ecologicalzone (an early form of the farm systemsrecommendation).

• Include a benefit-cost calculation in the assess-ment of irrigation projects.

The Second Evaluation included the fol-lowing main recommendations:• Adjust staffing patterns to more adequately

reflect changing workloads.• Improve the identification of women’s pro-

gram packages.• Strengthen staff resources in monitoring and

evaluation.• View savings and credit issues in the wider

context of total rural financial services in theNAC.

• Link with existing marketing channels inPakistan.

• Focus more on new technologies.• Emphasize improved animal nutrition over

increased herd size or the introduction ofnew breeds.

• Seek donor support for collaborative researchprojects.

• Expand the training function.

• Introduce labor savings packages for women.• Extend the Monitoring, Evaluation, and

Research Section beyond simply monitoring.• Codify an arbitration process at village level.• Cooperate with other agencies.• Establish a rural development academy.

The Third Evaluation did not discuss theextent to which the previous recommendations hadbeen implemented. However, the repetition ofearlier recommendations in the Third Evaluationsuggests some weaknesses in adoption and imple-mentation. The main recommendations of theThird Evaluation (not readily discerned from thereport since they were not separately listed) were:• Take a hard look at the role of the AKRSP in

development in the North.• Ensure that the savings and credit mechanisms

are sustained after the AKRSP.• Strengthen community organizations to

become semi-permanent entities.• Further fund productive and social infra-

structure.• Stimulate local entrepreneurial capacity.• Examine longer-term income prospects given

the increasing role of non-farm income.• Continue to be vigilant in monitoring pro-

grams and achievements in the poverty area—more targeted programs may be warranted.

• Consider the possibility of making commu-nity organizations legal or quasi-legal entities.

• Increase emphasis on the NRM in the nextphase.

• Identify different types of farming householdsand involve key farmers in project design andin evaluation of the results of experimentation.(The farm systems proposal.)

• Expand training in accounting and manage-ment.

ANNEX B: RESPONSIVENESS TO EARLIER OED RECOMMENDATIONS

• Focus on women of poor households, whoare often left out of women’s organizationmembership or activities.

The assessment of this Fourth Evaluation isthat, weighting by relative importance, about 75percent of the Third Evaluation recommendationshave been addressed to some significant extent.

Looking back to the earlier evaluations, a num-ber of recommendations are still valid today; forexample, the need to strengthen marketing link-ages, the farm systems recommendations in theirvarious forms, the need to address productivepackages for women, and the need for morelinkages with other agencies.

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

5 0

5 1

ANNEX C: AKRSP COST COMPARATORS

Tota

lA

nnua

lA

nnua

lpr

ojec

top

erat

ing

oper

atin

gTo

tal

cost

sPr

ojec

tco

sts

No.

of

$/co

st/

Coun

try/

proj

ect

Don

or($

m c

urre

nt)

year

s($

m)

hous

ehol

dsho

useh

old

hous

ehol

da,b

Com

pone

nts

Com

men

t

Paki

stan

Man

y$8

3.5m

17$3

.9m

(avg

. 93

,742

$42

$890

(17

yrs)

Com

mun

ity d

evel

opm

ent,

AKRS

P$3

6.1m

5ov

er 5

yrs

)$3

85 (5

yrs

)m

icro

finan

ce, a

gric

ultu

re

Sala

ries

64%

an

d liv

esto

ck p

rodu

ctio

n,

of c

ost (

excl

. fo

rest

ry, i

nfra

stru

ctur

e,

capi

tal)

ente

rpris

e de

velo

pmen

t,

train

ing

Paki

stan

Man

y +

own

$4.3

m

Cost

dat

a fo

r$2

.5m

avg

.28

4,48

5$9

Tota

l of

Com

mun

ity d

evel

opm

ent,

Com

para

ble

tota

l cos

ts

NRS

Pen

dow

men

tap

prox

.1

year

but

last

two

year

ssc

hem

es =

mic

rofin

ance

, agr

icul

ture

unce

rtain

due

par

tner

ship

s

2,00

0 on

lyru

nnin

g fo

r 7(in

cl. f

ield

appr

ox. $

11pr

oduc

tion,

infra

stru

ctur

e,

Habi

b Ba

nk, e

tc. A

bout

70

oper

atio

ns)

soci

alco

mm

unity

org

aniza

tions

per s

ocia

l mob

ilize

r

Paki

stan

Wor

ld B

ank

$32.

6m6

$4.0

m (e

st.)

2,60

0 di

rect

,$8

0+$6

52+

Agric

ultu

re a

nd li

vest

ock

pro-

ERR=

13%

Nor

ther

n Re

sour

ces

abou

t 50,

000

duct

ion,

fore

stry

, rur

al ro

ads,

Abou

t 12

com

mun

ity o

rgan

-

Man

agem

ent P

roje

ctro

ads,

land

wom

en in

dev

elop

men

t,iza

tions

per

socia

l mob

ilize

r

plan

com

mun

ity o

rgan

izatio

n,

land

pla

nnin

g, tr

aini

ng/T

A

Paki

stan

Pov

erty

W

orld

Ban

k$1

07m

5$2

.1m

200,

000

$10

$535

Mic

rofin

ance

, inf

rast

ruct

ure,

Us

es N

RSP,

AKRS

P, an

d

Alle

viat

ion

Fund

(but

ann

ually

($

52)

inst

itutio

n bu

ildin

g, tr

aini

ng,

othe

rs

abou

t 40,

000)

com

mun

ity d

evel

opm

ent

Oper

atin

g co

sts

diffi

cult

to

estim

ate

Paki

stan

IFAD

$25.

1m35

,600

$705

Agric

ultu

re, w

ells

, min

idam

s,N

RSP

will

impl

emen

t

Bara

nico

mm

unity

fund

, com

mun

ity

orga

niza

tion,

mic

roen

terp

rise

Paki

stan

IFAD

$22.

6m18

,000

$1,2

55

Nor

ther

n Ar

eas

Deve

lopm

ent

Co

st

C

om

pa

ra

to

rs

Ta

bl

e

C.

1

(con

tinue

d on

follo

win

g pa

ge)

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

5 2

Tota

lA

nnua

lA

nnua

lpr

ojec

top

erat

ing

oper

atin

gTo

tal

cost

sPr

ojec

tco

sts

No.

of

$/co

st/

Coun

try/

proj

ect

Don

or($

m c

urre

nt)

year

s($

m)

hous

ehol

dsho

useh

old

hous

ehol

da,b

Com

pone

nts

Com

men

t

Indi

a

AKRS

P (G

ujar

at)c

AKRS

PI$6

.71m

Over

5

$0.6

m50

7 co

mm

uni-

$19

$210

Com

mun

ity in

stitu

tions

, To

tal s

taff

151

1999

/00

ties—

abou

t ge

nder

, mic

rofin

ance

, NRM

,Sa

larie

s 74

% o

f ope

ratin

g

32,0

00

train

ing,

rese

arch

, pol

icy

cost

s

hous

ehol

dsad

voca

cy11

% fr

om o

wn

fund

ing

13%

wom

en m

embe

rshi

p

in 1

993

Burk

ina

IFAD

$25.

3m10

,000

$9

73+

Live

stoc

k w

ater

, soi

l con

ser-

Sout

hwes

t Rur

al

(dire

ct)

vatio

n, ro

ads/

track

s, c

redi

t,

Deve

lopm

ent

26,0

00so

cial

infra

stru

ctur

e, v

illag

e

(incl

udin

gw

ater

indi

rect

)

Bang

lade

shIFA

D$2

5.7m

82,0

00$3

13Re

sear

ch, w

ater

man

age-

Smal

lhol

der I

mp.

men

t, ru

ral i

nfra

stru

ctur

e

Ghan

aIFA

D$6

0m50

0,00

0$1

20M

arke

t inf

rast

ruct

ure,

rura

l

Villa

ge In

frast

ruct

ure

wat

er

Ethi

opia

IFAD

$20.

8m30

0,00

0$6

9In

stitu

tiona

l sup

port,

cre

dit,

Sout

hern

Reg

ion

rura

l roa

ds, s

eed

mul

tiplic

a-

Coop

erat

ive

tion,

trai

ning

Deve

lopm

ent

Nep

al

IFAD

$13.

1m25

,000

$524

Irrig

atio

n, fl

ood

reha

bilit

atio

n

Grou

ndw

ater

,

Irrig

atio

n &

Flo

od

Co

st

C

om

pa

ra

to

rs

(

co

nt

in

ue

d)

Ta

bl

e

C.

1

A n n e x e s

5 3

Indi

a W

orld

Ban

k$8

0.2m

7$1

.9m

but

80,0

00

$24

but

$1,0

00La

nd re

clam

atio

n, c

redi

t,

Sodi

c La

nds

excl

udes

ex

clud

esco

mm

unity

gro

ups,

ada

ptiv

e

Recl

amat

ion

Proj

ect

som

e co

sts

som

e co

sts

rese

arch

, ag.

ext

ensi

on,

train

ing

Mex

ico

Wor

ld B

ank

$115

.4m

10 m

illio

n$1

2Sm

all-s

cale

infra

stru

ctur

eOn

e co

mm

enta

tor s

ays

Solid

arity

Pro

gram

fund

s sp

read

too

thin

ly,

serio

usly

com

prom

ised

sust

aina

bilit

y

Braz

il W

orld

Ban

k$1

56m

208,

000

$749

Smal

l-sca

le in

frast

ruct

ure

Nor

thea

st R

ural

Deve

lopm

ent

Cear

a St

ate

Braz

il W

orld

Ban

k$9

6m11

2,00

0$8

56Sm

all-s

cale

infra

stru

ctur

e

Nor

thea

st R

ural

Deve

lopm

ent

Para

iba

Stat

ea.

IFAD

’s av

erag

e to

tal p

roje

ct c

ost i

s ab

out $

36 m

illio

n an

d av

erag

e ho

useh

olds

per

pro

ject

abo

ut 8

5,00

0, fo

r an

aver

age

$420

per

hou

seho

ld—

usua

lly o

ver a

bout

a 6

-yea

r pro

ject

per

iod.

b. A

ttrib

utin

g op

erat

ing

cost

s pe

r hou

seho

ld is

ver

y di

fficu

lt si

nce

in s

ome

prog

ram

s, s

uppo

rt is

for a

sho

rt pe

riod

to im

plem

ent s

peci

fic in

vest

men

ts o

f a o

ne-o

ff ki

nd, i

n ot

her p

rogr

ams,

cos

ts s

uppo

rt a

long

-term

rela

tions

hip

with

the

com

mun

ities

who

may

be

acce

ssin

g fu

ndin

g pa

rtly

from

insi

de th

e pr

ogra

m b

ut p

artly

from

out

side

.

c. In

form

atio

n pr

ovid

ed b

y AK

RSP

Indi

a.

5 5

ANNEX D: AKRSP TIMELINE

AKR

SP/C

omm

unity

O

rgan

izat

ions

Mic

rofin

ance

NRM

PPI

Gen

der

Ente

rpris

e de

velo

pmen

t

1981

Firs

t AKF

mee

ting

abou

t

initi

atin

g an

RDP

1982

Mr.

Shoa

ib S

ulta

n Kh

anCo

mm

unity

sav

ings

initi

ated

Fi

rst P

PI

hold

s fir

st c

omm

unity

dial

ogue

1983

Soci

al O

rgan

izatio

n Un

it Fi

rst s

hort-

term

loan

s—Ag

ricul

ture

and

live

stoc

kFi

rst w

omen

’s or

gani

zatio

nM

arke

ting

Sect

ion

form

ed

set u

pin

tere

st-fr

eeco

urse

s in

itiat

ed

Pak-

81 w

heat

intro

duce

d by

AKRS

P

1984

Med

ium

-term

loan

s—

Min

imum

sav

ings

leve

l set

8% in

tere

st

1985

DPO

Chitr

al; H

is H

ighn

ess

Mar

ketin

g co

urse

s in

itiat

ed

visi

ts fo

r Silv

er J

ubile

e

cele

brat

ions

1986

DPO

Skar

duDe

cent

raliz

ed m

anag

emen

t Se

para

te L

ives

tock

Sec

tion

Firs

t wom

en’s

grou

p w

ithin

Apric

ot d

ryin

g su

ppor

t

Firs

t OED

eva

luat

ion

of lo

ans

form

eda

villa

ge o

rgan

izatio

n in

in

itiat

ed

Thre

e-ye

ar fo

rest

ry p

ilot

Balti

stan

initi

ated

in G

ojal

with

IUCN

/ NOR

AD

1987

Core

Offi

ce b

ifurc

ated

, Fi

rst i

mpr

oved

bre

ed

Com

mer

cial

and

Indu

stria

l

DPO

Gilg

itan

imal

s fo

r col

lect

ive

Deve

lopm

ent D

ivis

ion

Acco

unts

cou

rses

initi

ated

man

agem

ent

form

ed

Firs

t qua

lity

appl

e pl

antin

g

mat

eria

l dis

tribu

ted

1988

Chitr

al A

rea

Deve

lopm

ent

Firs

t ser

vice

cha

rge

Appr

opria

te te

chno

logy

Se

cond

PPI

s ac

cept

ed, b

ut

Proj

ect (

IFAD/

ADB)

initi

ated

co

urse

s in

itiat

edin

new

vill

age

orga

niza

tions

mod

eled

on

AKRS

P

AK

RS

P

Ti

me

li

ne

Ta

bl

e

D.

1

(con

tinue

d on

follo

win

g pa

ge)

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

5 6

AKR

SP/C

omm

unity

O

rgan

izat

ions

Mic

rofin

ance

NRM

PPI

Gen

der

Ente

rpris

e de

velo

pmen

t

1989

Seco

nd O

ED e

valu

atio

nFi

rst g

roup

loan

s (v

illag

e Im

prov

ed b

reed

dis

tribu

tion

orga

niza

tion/

WCP

) gra

dual

lysw

itche

d to

indi

vidu

als

repl

aced

sho

rt-te

rm lo

ans

Fore

stry

cou

rses

initi

ated

1990

Dono

r Lia

ison

Gro

up F

orm

ed

Firs

t clu

ster

org

aniza

tion

(at A

wi i

n Bo

oni F

MU)

1991

Firs

t Far

m H

ouse

hold

Sur

vey

Villa

ge o

rgan

izatio

n

Firs

t Ins

titut

iona

l Mat

urity

co

oper

ativ

e m

arke

ting

Exer

cise

disc

ontin

ued

1992

Mr.

Shoa

ib S

ulta

n Kh

an

Ente

rpris

e Cr

edit

Prog

ram

Kail

shee

p in

trodu

ced

leav

es(E

CP) i

nitia

ted

(16%

)Ch

ick

broo

ding

cen

ters

Mr.

Huss

ain

Wal

i Kha

nes

tabl

ishe

d (n

ow p

rivat

ized)

take

s ov

er

NRS

P in

itiat

ed—

mod

eled

on A

KRSP

1993

Asto

re/D

iam

er D

istri

ct

adde

d

Firs

t don

or jo

int m

onito

ring

mis

sion

1994

Mr.

Stev

e Ra

smus

sen

take

s SD

C re

port

on R

ural

Ban

k

over

Seco

nd F

arm

Hou

seho

ld

Surv

ey

1995

Orga

niza

tiona

l cha

nge

Sepa

rate

sec

tion

split

off

Core

offi

ce d

owns

izing

Serv

ice

char

ges

rais

ed,

MER

Sec

tion

reor

gani

zed

loan

s fu

lly c

olla

tera

lized

FMUs

cre

ated

, gre

ater

dece

ntra

lizat

ion

AK

RS

P

Ti

me

li

ne

(

co

nt

in

ue

d)

Ta

bl

e

D.

1

A n n e x e s

5 7

1996

Form

al m

issi

on s

tate

men

t Co

rpor

ate

Cred

it Po

rtfol

io

Seco

nd P

PIs

acce

pted

deve

lope

din

itiat

edin

exi

stin

g vi

llage

Third

OED

eva

luat

ion

Firs

t pro

cedu

res

man

ual

orga

niza

tions

1997

Third

Far

m H

ouse

hold

St

arte

d en

terp

rise

deve

l-

Surv

eyop

men

t out

side

vill

age

orga

niza

tions

1998

Reac

hed

95%

sta

ffing

from

15

% V

OCP

annu

al s

ervi

ce

ED C

ente

r for

med

with

in th

e pr

ogra

m a

rea

char

ge

24%

ECP

1999

Disc

ontin

ued

shor

t-ter

m

loan

s

Med

ium

-term

loan

s

phas

ed o

ut d

ue to

poo

r

perfo

rman

ce—

mac

hine

ry

loan

s, e

tc.

V/W

OCP

rena

med

Gro

up

Loan

s

Corp

orat

e Cr

edit

disc

ontin

ued

2000

Dehi

Cou

ncils

intro

duce

d by

Ex

perim

entin

g w

ith

gove

rnm

ent

Busi

ness

Com

mitt

ee L

oans

Pove

rty p

ilots

initi

ated

in G

ilgit

tow

n

2001

Four

th O

ED e

valu

atio

n

5 9

A survey of professional staff (grades 4 andabove) reveals the AKRSP to be an effectiveorganization, with a healthy and broadly positiveinstitutional climate.1 The average score over allquestions was 4.0 out of a possible 5.0. Thus, witha few weaknesses noted below, the surveyreveals the AKRSP to be an organization staffedby people with a clear sense of purpose, a strongand shared commitment to its mission, the com-petence to implement its programs, and a com-mitment to monitor and learn from experience.

In surveys of this kind it is unusual to find noscores below the midpoint of 2.5—an inspectionof the table in this annex reveals that the low-est score for any question was 2.7. This suggeststhat answers may be biased upward across theboard and that respondents have chosen to dis-criminate in their judgments only within theupper part of the spectrum of answers.2 Suchresults are not unknown among populationswhere, as in the AKRSP, there has been little orno experience of this kind of enquiry.3 Thus, totease out meanings requires that more impor-tance be given to small differences in scores thanwould be the case in a survey with a more “nor-mal” spread of responses.

As table E.1 shows, the responses varied lit-tle by grade or regional office—a strong indi-cation of a highly homogenous institution. Thisallows attention to be concentrated on the resultsfor all respondents. Although the AKRSP’s direc-tion (mission and methods) is well understood,the link between personal performance andinstitutional goals is weaker than desired (items1.4, 1.6, and 1.8).

Staff clearly believe that the AKRSP has highstandards of openness and equity, and sharethese values themselves. However, there aresigns that these values are unevenly practiced(items 2.5, 2.6, and 2.7), suggesting the need forsome improvement in performance. Women aremore positive than men.

The aggregate response on the humanresource questions (section 3) is also lower thandesirable, which indicates that staff think somehuman resource policies are unfair and thatwomen are inadequately represented. The twoconcerns may well be interrelated. Similarly,although the group of questions on authority,responsibility, and accountability records a highscore overall, there are clear indications of pos-sible unfairness in the application of accounta-bility and in the reporting of bad news (items4.4 and 4.6). Management should consider anopen, focused, and well thought throughresponse to this possible unfairness.

The section on capability and competencereflects well on the AKRSP, but there are con-cerns about coordination and centralization(items 5.5 and 5.10). This may reflect current dis-cussion about the role and performance of FieldManagement Units (FMUs). It also reflects thetension inherent in spreading the Core Officeacross two locations—Gilgit and Islamabad.There is some concern about the AKRSP’s ten-dency toward introspection (item 3.6), but oth-erwise the AKRSP is clearly shown to be alearning organization (section 6) as well as beingopen and outward-facing (section 7).

ANNEX E: INSTITUTIONAL SURVEY

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

6 0

All

Wom

enM

enG

rd 4

Grd

5G

rd 6

Grd

7-9

Core

Ast

ore

B’st

anCh

itral

Gilg

it

1.

Dire

ctio

n4.

24.

44.

14.

34.

04.

14.

13.

94.

64.

24.

24.

1

1.1

I kno

w a

nd u

nder

stan

d AK

RSP’s

mis

sion

4.7

4.8

4.7

4.8

4.7

4.8

4.9

4.4

4.8

4.7

4.8

4.7

1.2

I kno

w a

nd u

nder

stan

d m

y ro

le in

fulfi

lling

AKR

SP’s

mis

sion

4.8

4.8

4.8

4.8

4.7

4.9

4.8

4.6

4.6

4.7

4.8

4.8

1.3

The

polic

ies

to a

chie

ve o

ur m

issi

on o

bjec

tives

are

cle

ar4.

24.

44.

14.

43.

84.

04.

23.

64.

84.

34.

24.

0

1.4

Polic

ies

to a

chie

ve o

ur m

issi

on o

bjec

tives

are

cle

arly

com

mun

icat

ed to

sta

ff at

all

leve

ls3.

73.

83.

63.

83.

43.

54.

03.

44.

03.

73.

63.

7

1.5

I und

erst

and

the

polic

ies

that

affe

ct m

y ro

le in

AKR

SP4.

24.

34.

24.

34.

04.

04.

34.

04.

74.

14.

34.

1

1.6

Our p

lans

are

fully

ade

quat

e to

ach

ieve

our

obj

ectiv

es3.

94.

23.

84.

13.

83.

63.

63.

64.

14.

23.

83.

7

1.7

Our p

lans

incl

ude

mea

sura

ble

perfo

rman

ce in

dica

tors

4.0

4.4

3.9

4.2

3.7

3.9

3.9

3.6

4.8

4.1

4.0

3.8

1.8

Perfo

rman

ce in

dica

tors

are

regu

larly

mea

sure

d an

d re

porte

d3.

94.

53.

74.

13.

64.

13.

33.

54.

84.

23.

83.

7

2.

Iden

tity

and

valu

es3.

94.

23.

83.

93.

73.

84.

03.

74.

33.

93.

73.

8

2.1

AKRS

P ad

voca

tes

high

sta

ndar

ds o

f ope

nnes

s, fa

irnes

s, a

nd

equi

ty4.

04.

43.

93.

93.

74.

44.

33.

64.

64.

13.

54.

1

2.2

AKRS

P cl

early

arti

cula

tes

the

stan

dard

s of

ope

nnes

s, fa

irnes

s,

and

equi

ty it

exp

ects

us

to p

ract

ice

3.9

4.2

3.9

3.9

3.9

4.1

4.0

3.7

4.2

4.0

3.7

4.0

2.3

I und

erst

and

and

shar

e th

e va

lues

adv

ocat

ed b

y AK

RSP

4.3

4.4

4.2

4.2

4.1

4.3

4.5

3.9

4.4

4.1

4.2

4.4

2.4

I pra

ctic

e th

ese

stan

dard

s an

d va

lues

4.3

4.6

4.2

4.3

4.1

4.2

4.3

4.2

4.6

4.3

4.2

4.2

2.5

Mos

t peo

ple

in A

KRSP

und

erst

and

and

shar

e its

sta

ndar

ds o

f

open

ness

, fai

rnes

s, a

nd e

quity

3.5

4.0

3.3

3.6

3.4

3.2

3.3

3.4

4.0

3.7

3.2

3.4

2.6

Mos

t peo

ple

in A

KRSP

pra

ctic

e its

sta

ndar

ds o

f ope

nnes

s,

fairn

ess,

and

equ

ity3.

43.

93.

33.

63.

33.

13.

43.

64.

03.

63.

33.

2

2.7

Peop

le in

AKR

SP a

re re

war

ded

fairl

y an

d in

acc

orda

nce

with

thes

e va

lues

3.

34.

03.

23.

43.

02.

94.

53.

14.

13.

62.

93.

3

2.8

On b

alan

ce, l

evel

s of

pro

fess

iona

l tru

st in

AKR

SP a

re h

igh

3.9

4.2

3.8

4.0

3.6

4.0

4.0

3.6

4.2

3.8

3.9

3.8

2.9

I tru

st th

e pr

ofes

sion

al ju

dgm

ents

of m

y co

lleag

ues

3.9

4.1

3.8

4.0

3.7

4.1

3.9

3.7

4.5

3.7

3.8

3.9

2.10

Ther

e is

a s

trong

sen

se o

f tea

mw

ork

in A

KRSP

4.1

4.3

4.1

4.3

4.0

4.0

4.0

3.7

3.9

4.3

4.3

4.0

3. H

uman

reso

urce

s3.

64.

13.

53.

73.

43.

53.

53.

54.

03.

83.

43.

5

3.1

AKRS

P’s h

uman

reso

urce

pol

icie

s ar

e fra

med

and

ope

rate

d in

acco

rdan

ce w

ith it

s va

lues

of o

penn

ess,

fairn

ess,

and

equ

ity3.

54.

13.

43.

73.

33.

33.

63.

44.

13.

73.

33.

5

AK

RS

P

In

st

it

ut

io

na

l

Su

rv

ey

Ta

bl

e

E.

1

A n n e x e s

6 1

3.2

On b

alan

ce th

e st

aff i

n AK

RSP

are

wel

l qua

lifie

d to

ach

ieve

its

mis

sion

and

obj

ectiv

es4.

14.

54.

04.

14.

04.

04.

14.

14.

44.

13.

94.

0

3.3

On b

alan

ce A

KRSP

’s hu

man

reso

urce

pol

icie

s ar

e w

ell a

djus

ted

to th

e ac

hiev

emen

t of i

ts m

issi

on3.

64.

13.

53.

73.

53.

43.

73.

54.

23.

83.

43.

5

3.4

The

gend

er b

alan

ce a

mon

gst s

taff

in A

KRSP

is a

bout

righ

t3.

13.

63.

03.

22.

93.

22.

72.

93.

23.

42.

93.

1

4.

Auth

ority

, res

pons

ibili

ty, a

nd a

ccou

ntab

ility

4.0

4.4

3.9

4.1

3.8

3.8

4.0

3.8

4.6

4.1

3.8

3.9

4.1

Key

deci

sion

s in

AKR

SP a

re m

ade

by p

eopl

e w

ith a

dequ

ate

expe

rtise

and

aut

horit

y3.

94.

43.

84.

04.

03.

93.

84.

14.

83.

93.

73.

8

4.2

Mos

t peo

ple

in A

KRSP

hav

e en

ough

aut

horit

y an

d re

spon

sibi

lity

to d

o th

eir j

obs

prop

erly

4.1

4.4

4.0

4.3

3.9

3.7

4.1

3.9

4.7

4.2

4.0

4.0

4.3

I hav

e en

ough

aut

horit

y an

d re

spon

sibi

lity

to d

o m

y jo

b pr

oper

ly4.

34.

64.

24.

53.

93.

84.

44.

14.

74.

54.

24.

0

4.4

Mos

t peo

ple

in A

KRSP

, inc

ludi

ng th

e m

ost s

enio

r, ar

e he

ld

acco

unta

ble

for t

heir

actio

ns a

nd d

ecis

ions

3.6

4.2

3.5

3.8

3.4

3.7

3.6

3.3

4.2

3.8

3.3

3.7

4.5

I am

hel

d ac

coun

tabl

e fo

r my

actio

ns a

nd d

ecis

ions

4.3

4.6

4.3

4.5

4.2

4.0

4.3

4.1

4.7

4.4

4.4

4.2

4.6

Mis

take

s, fa

ilure

s, a

nd b

ad n

ews

are

repo

rted

to th

ose

who

need

to k

now

with

out f

ear o

f rep

risal

3.7

4.1

3.5

3.7

3.6

3.5

3.7

3.5

4.8

3.5

3.5

3.6

5.

Capa

bilit

y an

d co

mpe

tenc

e4.

14.

44.

04.

23.

94.

14.

14.

04.

44.

23.

94.

1

5.1

On b

alan

ce A

KRSP

is a

com

pete

nt o

rgan

izatio

n4.

74.

84.

64.

74.

64.

94.

74.

64.

84.

64.

54.

8

5.2

Mos

t peo

ple

in A

KRSP

hav

e th

e kn

owle

dge

and

skill

s to

do

thei

r job

wel

l4.

24.

54.

14.

24.

24.

14.

24.

24.

34.

34.

04.

2

5.3

Mos

t peo

ple

in A

KRSP

hav

e th

e to

ols

and

reso

urce

s to

do

thei

r

job

wel

l4.

24.

74.

14.

44.

04.

04.

04.

14.

64.

44.

24.

0

5.4

Proc

edur

es a

nd p

roce

sses

in A

KRSP

are

wel

l tho

ught

out

and

effic

ient

4.0

4.3

3.9

4.1

3.7

4.2

3.9

3.7

4.3

4.0

3.8

4.0

5.5

Diffe

rent

sec

tions

/dep

artm

ents

are

wel

l coo

rdin

ated

with

one

anot

her

3.6

3.8

3.6

3.7

3.4

3.7

3.7

3.3

3.7

3.6

3.6

3.6

5.6

My

perfo

rman

ce is

regu

larly

mon

itore

d an

d I a

m g

iven

cons

truct

ive

feed

back

4.1

4.2

4.0

3.9

4.2

4.5

4.0

4.4

4.2

3.7

3.9

4.2

5.7

I hav

e th

e to

ols

and

reso

urce

s to

do

my

job

wel

l 4.

24.

54.

14.

34.

04.

24.

04.

24.

84.

33.

94.

1

5.8

I hav

e th

e sk

ills

and

train

ing

to d

o m

y jo

b w

ell

4.2

4.6

4.1

4.4

4.0

4.0

4.2

3.9

4.7

4.3

4.2

4.1

5.9

AKRS

P’s o

rgan

izatio

nal s

truct

ure

is w

ell s

uite

d to

enh

anci

ng

its o

bjec

tives

4.2

4.6

4.1

4.3

3.9

4.3

4.6

4.3

4.7

4.4

3.9

4.2

(con

tinue

d on

follo

win

g pa

ge)

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

6 2

All

Wom

enM

enG

rd 4

Grd

5G

rd 6

Grd

7-9

Core

Ast

ore

B’st

anCh

itral

Gilg

it

5.10

The

bala

nce

betw

een

cent

raliz

atio

n an

d de

cent

raliz

atio

n in

AKRS

P is

abo

ut ri

ght

3.6

4.3

3.5

3.7

3.5

3.4

4.1

3.4

4.4

3.9

3.5

3.5

6.

Mon

itorin

g an

d le

arni

ng4.

14.

24.

14.

24.

04.

14.

14.

04.

24.

14.

14.

1

6.1

AKRS

P re

gula

rly re

view

s th

e ex

tern

al a

nd in

tern

al e

nviro

nmen

ts

to s

ee if

cha

nges

in m

issi

on, g

oals

, or p

olic

ies

are

need

ed4.

24.

64.

14.

34.

04.

24.

33.

94.

84.

34.

04.

2

6.2

AKRS

P re

gula

rly m

onito

rs it

s pe

rform

ance

aga

inst

rele

vant

indi

cato

rs/b

ench

mar

ks4.

24.

64.

24.

34.

14.

04.

34.

14.

74.

34.

24.

1

6.3

AKRS

P sp

ends

too

muc

h tim

e an

d re

sour

ces

revi

ewin

g its

elf

and

its s

trate

gy a

nd p

olic

ies

3.6

3.3

3.7

3.5

3.9

3.6

3.6

3.6

3.1

3.4

3.7

3.8

6.4

AKRS

P is

a le

arni

ng o

rgan

izatio

n, d

raw

ing

less

ons

from

expe

rienc

e &

mak

ing

posi

tive

polic

y &

org

aniza

tiona

l cha

nges

4.4

4.5

4.4

4.5

4.2

4.4

4.3

4.2

4.3

4.4

4.5

4.3

6.5

Peop

le in

AKR

SP a

re e

ncou

rage

d to

avo

id w

aste

and

be

effic

ient

4.1

4.2

4.1

4.2

4.0

4.1

4.3

3.9

4.3

4.1

4.3

4.0

7.Ex

tern

al re

latio

ns4.

34.

34.

34.

34.

24.

34.

24.

04.

14.

24.

34.

3

7.1

AKRS

P ha

s an

act

ive

and

wid

e-ra

ngin

g pr

ogra

m o

f ext

erna

l

rela

tions

insi

de a

nd o

utsi

de P

akis

tan

4.4

4.4

4.4

4.4

4.3

4.3

4.2

4.0

4.3

4.4

4.4

4.3

7.2

AKRS

P ha

s m

any

prog

ram

s th

at d

epen

d on

ext

erna

l par

tner

s4.

14.

04.

14.

14.

24.

23.

94.

43.

64.

04.

24.

1

7.3

AKRS

P sh

ould

see

k to

exp

and

its p

rogr

ams

with

ext

erna

l

partn

ers

4.2

4.1

4.3

4.2

4.2

4.2

4.5

3.8

3.9

4.1

4.3

4.4

7.4

Exte

rnal

par

tner

s an

d ot

her d

onor

s ha

ve a

gen

eral

ly p

ositi

ve

influ

ence

on

AKRS

P4.

44.

54.

44.

44.

44.

34.

34.

04.

34.

34.

64.

4

7.5

AKRS

P ha

s m

utua

lly b

enef

icia

l lin

ks w

ith g

over

nmen

t age

ncie

s 4.

14.

34.

04.

23.

94.

33.

93.

74.

44.

14.

04.

1

7.6

AKRS

P sh

ould

exp

and

and

deep

en it

s lin

ks w

ith g

over

nmen

t

agen

cies

4.4

4.4

4.4

4.4

4.3

4.2

4.6

4.1

4.2

4.4

4.4

4.3

Max

imum

num

ber o

f obs

erva

tions

170

3013

983

4321

2014

939

4663

AK

RS

P

In

st

it

ut

io

na

l

Su

rv

ey

(

co

nt

in

ue

d)

Ta

bl

e

E.

1

6 3

Gilgit RegionMurtazabadMominabadKhaiberTholeGhulmitBudalasTeruGahkuch BalaKakhan, AstoreAliabad, Astore

Baltistan RegionStakShagri KalanChurkaThugmoDaghoniYuguRangaHoto

Chitral Region Krui Jinali BooniYarkhoon LashtBaranisKaghoziKalkatakBirirShagramYorjogh

In all but 2 of the 26 villages on this list, the teamapplied four semi-formal questionnaires on com-munity organizations, natural resource man-agement, infrastructure, and microfinance. Threemission members spent about two-and-a-halfhours at each village, usually about one-and-a-

half hours with the full group, then about one-and-a-half hours in breakout groups and visit-ing infrastructure sites. In nearly all cases thewomen members of the team interviewedwomen’s groups separately.

ANNEX F: VILLAGES VISITED

6 5

ANNEX G: COMPUTATION OF SUBSIDY DEPENDENCE INDEX FOR THEMICROFINANCE OPERATIONS OF THE AKRSP

Alternative methodology

Particulars 1994 1997 1998 1999 2000 (2000)

S=A*(m-c)+[(E*m)—P] +K

S= Annual subsidy received by AKRSP for credit

operations, derived as follows:

A= AKRSP concessional borrowed funds outstanding

to fund the credit portfolio Rs. million 0 0 0 0 0 58.8a

m= Interest rate that AKRSP would have to pay for funds

if it could no longer obtain concessional funds 12% 12% 10% 6% 6% 12%

c= Interest rate paid by AKRSP on concessional borrowed

fund outstanding 0% 0% 0% 0% 6%b

E= Average annual equity or net worth of the credit

operations Rs. million 200.0 450.7 495.1 517.7 601.3 601.3

P= Reported annual profit (not adjusted for inflation) Rs. million 9.1 26.7 44.5 41.4 78.6 78.6

K= Sum of all other types of annual subsidies received by

AKRSP for its credit operations Rs. million 0 0 0 0 0 0

S= 14.9 27.4 5.0 (10.3) (42.5) (3)

SDI=S/(LP*n)

SDI= Subsidy dependence index, derived as

LP= Average annual outstanding portfolio Rs. million 127.0 241.0 263.4 227.0 171.5 171.5

n= The rate of interest or service charge on AKRSP loans 7.0% 10.5% 13.1% 15.3% 19.1% 19.1%

SDI= 1.7 1.1 0.1 (0.3) (1.3) (0.1)

or 168% 108% 15% -30% -130% -9%

RR=n*(SDI+1)

RR= Required rate of service charges to eliminate subsidy 18.7% 21.9% 15.0% 10.7% -5.7% 17.3%Note: The Alternative Methodology column takes a 12% opportunity cost of capital and adds in the Pakistan Poverty Alleviation Fund [PPAF] borrowing:

a. Balance of short-term borrowing from Pakistan Poverty Alleviation Fund (PPAF) at the end of 2000.

b. Service charge payable to PPAF at the rate of 6 percent per annum.

Source: AKRSP Microfinance Operations Financial Statements as of December 31, 2000.

6 7

ANNEX H: AKRSP ECONOMIC ANALYSIS

Item

Sele

cted

yea

rs (c

olum

ns h

idde

n fo

r pre

sent

atio

n):

1982

–83

1984

1990

1996

2001

2039

1Ca

pita

l exp

endi

ture

(rup

ees

’000

s)1,

931

1,99

25,

002

14,8

04

2To

tal o

pera

ting

cost

(rup

ees

’000

s) a

djus

t. by

SCF

4,39

65,

479

44,3

8714

2,74

9–

–

3Tr

aini

ng (r

upee

s ’0

00s)

296

913

8,94

417

,596

4Re

sear

ch, s

urve

y, an

d de

mon

stra

tion

(rupe

es ’0

00s)

1,72

63,

289

9,66

331

,493

5In

frast

ruct

ure

proj

ects

(rup

ees

’000

)7,

204

12,0

8335

,928

33,0

25

6Cr

edit

prog

ram

(net

) (ru

pees

’000

s)51

715

32,

290

43,4

59

7To

tal n

on-o

pera

ting

cost

(rup

ees

’000

)9,

743

16,4

3856

,825

125,

573

––

8To

tal c

osts

(rup

ees

’000

)14

,627

22,5

2610

6,14

428

4,18

3

9To

tal c

osts

with

eco

n. a

djus

t (in

199

9 ru

pees

’000

)49

,582

69,6

9122

8,60

332

5,82

4

10GD

P de

flato

r (ba

se 1

981

= 10

0)11

512

618

134

0

11Fa

ctor

to b

ring

to 1

999

rupe

es3.

83.

52.

41.

31.

01.

0

12Hy

del m

odel

net

stre

am (R

s) w

orst

cas

e-1

,000

.010

0.0

270.

027

0.0

270.

0

13Hy

del m

odel

net

stre

am IR

R22

%

14Hy

del w

eigh

t fac

tor (

incl

udin

g “o

ther

”)0.

33

15La

nd d

evel

opm

ent n

et s

tream

wor

st c

ase

-1,0

00.0

0.0

300.

040

0.0

400.

0

16La

nd d

evel

opm

ent n

et s

tream

IRR

19%

17La

nd d

evel

opm

ent w

eigh

t fac

tor

0.44

18Co

mm

unic

atio

ns n

et s

tream

wor

st c

ase

-1,0

00.0

0.0

300.

040

0.0

400.

0

19Co

mm

unic

atio

ns n

et s

tream

IRR

19%

20Co

mm

unic

atio

ns w

eigh

t fac

tor

0.23

21W

eigh

ted

mod

el n

et s

tream

-1,0

0032

284

350

350

22Ag

greg

ate

mod

el s

tream

IRR

19%

23Ab

ando

ned

proj

ect f

acto

r (4%

@ 5

0% e

xp.)

0.98

0.98

0.98

0.98

0.98

24Vi

llage

con

tribu

tion

fact

or+e

con

adju

stm

ent (

30%

)1.

301.

301.

301.

301.

30

2519

83 In

vest

—m

odel

flow

(1,0

00)

3228

435

035

0–

2619

83 A

ggeg

ate

net i

nfra

stru

ctur

e st

ream

(inc

l. 30

% v

ill. c

ont.)

-35,

669,

197

1,15

3,54

210

,140

,681

12,4

82,7

2112

,482

,721

0

2719

84 In

vest

—m

odel

flow

-1,0

0024

135

035

00

2819

84 A

ggre

gate

net

infra

stru

ctur

e st

ream

-54,

603,

652

13,1

69,2

0019

,108

,985

19,1

08,9

85

2919

85 In

vest

—m

odel

flow

219

350

350

0

AK

RS

P

Ec

on

om

ic

A

na

ly

si

sT

ab

le

H

.1

(con

tinue

d on

follo

win

g pa

ge)

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

6 8

Item

Sele

cted

yea

rs (c

olum

ns h

idde

n fo

r pre

sent

atio

n):

1982

–83

1984

1990

1996

2001

2039

3019

85 A

ggeg

ate

net i

nfra

stru

ctur

e st

ream

12,8

87,8

0120

,628

,569

20,6

28,5

69

3119

86 In

vest

—m

odel

flow

219

327

350

0

3219

86 A

ggeg

ate

net i

nfra

stru

ctur

e st

ream

14,1

79,3

2721

,234

,035

22,6

95,8

21

3319

94 A

ggre

gate

net

infra

stru

ctur

e st

ream

2,57

8,81

611

,335

,069

3419

95 In

vest

—m

odel

flow

3232

70

3519

95 A

ggre

gate

net

infra

stru

ctur

e st

ream

1,86

3,78

318

,869

,393

3619

96 In

vest

—m

odel

flow

-1,0

0021

90

3719

96 A

ggre

gate

net

infra

stru

ctur

e st

ream

-55,

307,

162

12,0

92,2

47

3819

97 In

vest

—m

odel

flow

219

0

3919

97 A

ggre

gate

net

infra

stru

ctur

e st

ream

19,1

93,5

06

4019

98 In

vest

—m

odel

flow

110

0

4119

98 A

ggre

gate

net

infra

stru

ctur

e st

ream

12,0

35,2

25

4219

99 In

vest

—m

odel

flow

6526

3

4319

99 A

ggre

gate

net

infra

stru

ctur

e st

ream

7,08

7,77

7

44N

et b

enef

it st

ream

infra

stru

ctur

e (a

fter e

con

adju

st)

(35,

669,

197)

(53,

450,

111)

(44,

446,

910)

171,

365,

232

355,

338,

314

28,7

80,6

69

45IR

R in

frast

ruct

ure

only

19%

46N

umbe

r of h

ouse

hold

s (h

h)15

,449

30,6

1262

,360

83,2

9494

,422

47Av

erag

e hh

. far

m in

clud

ing

extra

pola

tion

from

sur

veys

(199

9 ru

pees

)20

,270

20,6

8423

,349

62,1

0675

,515

48Av

erag

e pr

ogra

m to

tal f

arm

inco

me

313,

154,

393

633,

175,

007

1,45

6,06

8,18

05,

173,

049,

815

7,13

0,23

0,57

30

49To

tal f

arm

inco

me

incr

emen

t (@

10%

attr

ib.)+

econ

. adj

ust.

028

,481

,835

101,

719,

327

432,

530,

693

606,

719,

780

0

50Av

erag

e of

f-far

m h

h in

c. e

xtra

p. fr

om s

urve

ys (1

999

rupe

es)

37,3

2638

,087

42,9

9645

,848

48,1

23

51Av

erag

e pr

ogra

m to

tal o

ff-fa

rm in

com

e57

6,64

2,05

31,

165,

927,

556

2,68

1,20

1,87

43,

818,

863,

312

4,54

3,82

8,56

40

52To

tal o

ff-fa

rm in

com

e in

crem

ent (

@5%

attr

ib.)+

econ

. adj

ust.

026

,223

,205

93,6

52,9

1214

4,27

8,84

617

6,53

9,80

00

53To

tal N

RM c

osts

afte

r eco

n. a

djus

t.6,

047,

904

10,5

18,5

35.2

423

,418

,413

.05

68,4

93,2

47.5

315

0,13

4,61

4.6

54To

tal N

RM b

enef

its a

fter e

con.

adj

ustm

ent a

djus

t dow

n by

33%

00

19,2

07,9

89.1

611

5,33

3,15

427

5,38

9,63

0.7

55N

et b

enef

its N

RM a

fter e

con.

adj

ust.

-6,0

47,9

04-1

0,51

8,53

5.24

-4,2

10,4

23.8

9246

,839

,906

.45

125,

255,

016.

1

56IR

R N

RM o

nly

24%

57To

tal o

pera

ting

cost

s+ec

on. a

dj. (

’99

rupe

es, 2

000

to 2

009

at 5

0%)

16,7

41,5

0319

,046

,743

107,

411,

877

183,

894,

300

93,7

67,8

50

58To

tal c

osts

+eco

n. a

dj. (

’99

rupe

es)

49,5

81,7

1469

,691

,153

228,

602,

509

325,

824,

168

93,7

67,8

50

AK

RS

P

Ec

on

om

ic

A

na

ly

si

s

(c

on

ti

nu

ed

)T

ab

le

H

.1

A n n e x e s

6 9

59N

et in

frst.

+ N

RM +

5%

non

farm

ben

.-tot

. cos

t (-R

&D

& in

frst.)

(6

1,02

8,44

4)(5

9,87

8,55

4)(8

5,41

7,51

9)11

0,63

1,60

156

3,36

5,27

928

,780

,669

60ER

R of

abo

ve s

tream

16%

61N

et in

frst.

+ 10

% n

et fm

. inc

. + 5

% n

onfa

rm b

en.-t

ot. c

ost (

-R&

D &

infrs

t.)-5

4,98

0,54

0-2

0,87

8,18

420

,512

,232

496,

322,

387

1,04

4,83

0,04

328

,780

,669

62ER

R of

abo

ve s

tream

33%

1,99

25,

002

14,8

04N

ote:

Row

s 1

to 9

: AKR

SP c

osts

and

tota

ls w

ith S

tand

ard

Conv

ersi

on F

acto

r (SC

F) e

cono

mic

adj

ustm

ents

. (SC

F ad

just

s fin

anci

al fi

gure

s fo

r eco

nom

ic d

isto

rtion

s in

the

econ

omy)

Row

s 12

to 2

0: s

impl

ified

mod

el n

et b

enef

it st

ream

s de

rived

from

AKR

SP c

ase

stud

ies

usin

g w

orst

-cas

e sc

enar

ios.

Row

21:

wei

ghte

d m

odel

net

stre

am w

eigh

ted

by s

hare

of e

ach

type

of i

nfra

stru

ctur

e in

tota

l.

Row

23:

fact

or to

allo

w fo

r 4 p

erce

nt a

band

oned

pro

ject

s sp

endi

ng 5

0 pe

rcen

t of e

xpec

ted

cost

s.

Row

24:

to fa

ctor

to a

dd th

e vi

llage

con

tribu

tion

not i

nclu

ded

in c

osts

.

Row

s 25

to 4

3: a

ggre

gatio

n of

ann

ual i

nfra

stru

ctur

e m

odel

net

ben

efit

stre

ams.

Row

s 44

and

45:

tota

l of a

ll an

nual

stre

ams

and

Econ

omic

Rat

e of

Ret

urn.

Row

47:

ave

rage

hou

seho

ld fa

rm in

com

e ta

ken

from

the

1991

and

199

7 So

cio-

Econ

omic

Sur

veys

pro

ject

ing

forw

ard

and

back

by

an a

ssum

ed 2

per

cent

ann

ually

.

Row

48:

num

ber o

f hou

seho

lds

times

farm

inco

me.

Row

49:

incr

emen

tal t

otal

farm

inco

me

incr

ease

s as

sum

ing

10 p

erce

nt is

attr

ibut

able

to A

KRSP

and

app

lyin

g ec

onom

ic a

djus

tmen

ts.

Row

s 50

to 5

2: m

etho

dolo

gy a

s fo

r far

m in

com

e ab

ove

but t

akin

g th

e no

n-fa

rm in

com

e el

emen

t fro

m th

e So

cio-

Econ

omic

Sur

veys

and

attr

ibut

ing

5 pe

rcen

t to

AKRS

P.

Row

s 53

and

54:

take

n fro

m A

KRSP

NRM

eco

nom

ic a

naly

sis

with

eco

nom

ic a

djus

tmen

ts a

nd a

djus

ting

bene

fit s

tream

dow

n by

33

perc

ent t

o al

low

for s

ome

optim

ism

in a

ssum

ptio

ns.

Row

s 57

and

58:

tota

l AKR

SP o

pera

ting

cost

s an

d to

tal c

osts

+ e

cono

mic

adj

ustm

ents

; mai

nten

ance

leve

l of o

pera

ting

cost

s as

sum

ed to

yea

r 200

9 at

50

perc

ent o

f mos

t rec

ent.

Row

59:

net

infra

stru

ctur

e be

nefit

stre

am p

lus n

et n

atur

al re

sour

ce m

anag

emen

t ben

efit

stre

am p

lus f

ive

perc

ent a

ttrib

utio

n of

est

imat

ed n

on-fa

rm in

com

e de

rived

from

surv

eys d

educ

ting

tota

l AKR

SP co

sts w

ith a

djus

tmen

t for

thos

e co

sts a

lread

y

acco

unte

d fo

r the

in th

e in

frast

ruct

ure

and

NRM

net

ben

efit

stre

ams.

Row

61:

sim

ilar t

o ro

w 5

9 bu

t ins

tead

of A

KRSP

NRM

net

ben

efit

stre

am u

sing

the

assu

mpt

ion

of 1

0 pe

rcen

t of t

he a

ggre

gate

net

farm

inco

me

attri

buta

ble

to A

KRSP

bas

ed o

n ex

trapo

latio

n fro

m S

ocio

econ

omic

Sur

veys

. Wou

ld in

clud

e so

me

degr

ee

of d

oubl

e co

untin

g.

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

7 0

Based on Natural Resource Management Based on attributed share of total farm Methodology net benefit projections income increments

AKRSP base case(“best bet” Worst case

infrastructure and infrastructure andno 33% 33% reduction in Attributing Attributing

NRM reduction) NRM benefits 10% to AKRSP 20% to AKRSP

Attributing 5% non-farm benefits 24% 16%b 33% 58%

Attributing zero non-farm benefits 17% 11% 22% 38%

Attributing 5% non-farm benefits but with

investment and operating costs up 20%a 19% 12% 30% Not estimateda. While this gives an indication of the sensitivity to costs, it is not a realistic scenario since, with 17 years of cost data in the analysis, a large share of costs are actuals not projections

(operating costs are assumed to continue for 10 years at half the final level).

b. Deducting an estimate of the micro-finance costs from this scenario (assuming they are 10% of credit disbursements since these were not separately accounted for until recently) changes

the ERR by less than 1%.

S e n s i t i v i t y A n a l y s i sT a b l e H . 2

Main Enterprise Assumptions in Economic AnalysisInfrastructure assumptions are in Synthesis ofthe Findings from Impact Studies (AKRSP 2000o),and in AKRSP documents on CommunicationsProjects (AKRSP 2000f), Land Development Pro-jects (AKRSP 2000d), and Power Generation

(AKRSP 2000e). Assumptions in the latter threereports were taken partly from case study villageinterviews and partly from generic assumptions,but they are similar to those below.

A n n e x e s

7 1

Cost of improved Unimproved Improved (kg/kanal seed as % of

Enterprise (kg/kanal) or value) revenue or actual Comment

Wheat grain 75 112/125/141 (varieties) 4% 10 years to grow wheat on new land—

Border price of Rs9.3/kg fodder first

used New and old land treated differently

90% adoption of improved seed

provided

Wheat straw 262 213/230/309 (varieties)

Maize grain 107 150 4%

Maize straw 750 1,100

Potato 900 500 5% (seed rate 100kg@Rs8)

Alfalfa (dry) 850 1,450 2% 70% on new land

Fuelwood 40kg/tree@Rs2.7 Costs 63% of benefits After ’96: Poplar, 50%

Robina, 16%

Russian Olive, 19%

Willow, 15%

Timber Rs.2,000/tree or Rs5/kg Costs 50% of benefits

varies by species

Fruit (e.g., apple) 80kg/tree@Rs30

(apricot dried, Rs20) Cost 8% of revenue 80% on exiting land

Apple 88% before ’92, 45% after

Wastage 10%

Alternate bearing allowed for

Yield start year 5

Cattle: calves/yr 0.5

Milk (incremental) 5.3lt/day@ Rs18/lt

Sheep (incremental 19kg@Rs75 Rs.1,000 annual maintenance

meat) cf. local

Sheep (incremental 1kg/day

milk)

Medication impact Assumed 50% treated saved

Vaccination mortality Cattle saved 4%

impact Sheep saved 2%Note: An overall Standard Conversion Factor economic adjustment of 0.89 was used and commodity-specific adjustments for some traded commodities.

M a i n S e l e c t e d A s s u m p t i o n s f r o m A K R S PN R M A n a l y s i s

T a b l e H . 3

7 3

KARINA, established in 1984 at Jaglote on theKarakoram Highway, has the mandate to con-duct and coordinate applied research for theNorthern Areas (it does not cover Chitral) on fieldcrops, fruits and vegetables, livestock, fisheries,land, and water, and to transfer the findings tofarmers.

After many years of relative inactivity, KARINAnow has new leadership and a much-improvedprogram. It currently has 11 scientists, led by adirector, and an annual nondevelopment budgetof about Rs. 4 million (average over 1997 and2001). Having established a scientific staff andobtained almost 1.5 percent of the Pakistan Agri-culture Research Council’s nondevelopmentbudget, it is poised to play a greater role inenhancing productivity in the Northern Areas.The institute has five stations, four of themlocated in the Northern Areas (Skardu, Ghizer,Astore, and Chilas). The fifth, a trout multipli-cation and research station, was recently estab-lished in Kaghan as a collaborative partnershipwith the AKRSP.

While in recent years the impact of KARINAat the village level has been modest, about 1,000lines and cultivars of major wheat varieties havebeen tested over the past five years. Two culti-vars of wheat, Chakwal-86 and Chakwal-97, arebeing promoted in areas near KARINA and in theDiamer District. In 2001, KARINA had morethan 750 lines of wheat of more than a dozenvarieties in observation trials for seed purityand disease resistance. In 2000, a mass selectionof 80 lines was done and, based on their yieldperformance, disease resistance, and grain-strawratio, KARINA has selected 20 “better” per-

formers for further observation and will release3 to 5 wheat varieties in 2002.

KARINA has also established observation andadaptive lines of 7 main varieties of maize(Agaiti-85, Azam, Kisan, Pahar, a few hybrids),and vegetables and fodder (legumes). It plansto produce about 4 metric tons each of wheatand maize seed during 2001 for further multi-plication in farmers’ fields. During the preced-ing year, KARINA produced about 2.4 metric tonsof wheat and 1.6 metric tons of maize seed, dis-tributed to about 200 farmers, mainly from Chi-las and Diamer—mostly those who approachedKARINA. A promising commercial cultivar ofmuskmelon (Ravi) has been developed for thedouble-cropping zone. A successful cultivationtechnique has been developed for two medici-nal plants (black zera and kuth). They havealso collected, tested, and introduced local andexotic varieties of fruits, but this is little knownto farmers. KARINA’s progeny orchards have acollection of about 12 deciduous and other fruittree varieties: apples, apricots, pears, cherries,peaches, plums, almonds, grapes, pomegranates,figs, and olives. Work on rootstock resistance tocrown gall for cherry trees is under way.

While KARINA’s work program covers manyrelevant activities, it needs a better-articulatedstrategy and the resources to deliver its findingsto farmers and to interact with them. It needs atechnology transfer unit that, among other things,should promote effective partnerships with localinstitutions such as the AKRSP as vehicles toreach a widespread farming community. Thereis a need for increased institutional linkagesand a strategy forum.

ANNEX I: THE KARAKORAM AGRICULTURAL RESEARCH INSTITUTE FORTHE NORTHERN AREAS (KARINA)

7 5

ANNEX J: STATISTICAL TABLES

Program area Gilgita Chitral Baltistan Total

Social organization

Organizations (number) 1,460 1,140 1,168 3,768

Village organizations 798 758 815 2,371

Women’s organizations 662 382 353 1,397

Organization members (number) 62,229 39,706 39,542 141,477

Village organizations 35,485 29,116 29,821 94,422

Women’s organizations 26,744 10,590 9,721 47,055

Total savings (Rs. million) 269.86 94.64 65.75 430.25

Village organizations 182.32 76.82 60.50 319.64

Women’s organizations 87.54 17.82 5.25 110.61

Women in development

Credit disbursedb (Rs. million) 156.75 27.00 7.36 191.10

Group loans 1,256 472 120 1,848

Beneficiary households 54,347 12,132 3,794 70,273

WO specialists (number) 2,984 2,958 1,804 7,746

Productive infrastructure and engineering services

PPIs initiated (number) 638 779 802 2,219

Beneficiary households (number) 64,996 49,215 52,524 166,735

Cost of initiated PPIs (Rs. million) 217.75 304.45 245.74 767.94

PPIs completed (number) 515 720 682 1,917

Physical progress (% completed) 81 92 85 86

Agriculture

Improved seeds (kgs) 253,752 121,565 514,254 889,571

Total fruit trees (millions) 2.05 0.56 1.36 3.97

VO/WO agriculture specialists (number) 1,535 1,418 835 3,788a. Includes Astore.

b. Includes short-term, medium-term, and women’s organization credit.

A c h i e v e m e n t s i n P r o g r a m A r e a s , D e c e m b e r 1 9 8 2 – D e c e m b e r 2 0 0 0

T a b l e J . 1

The tables in this Annex were supplied by the AKRSP.

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

7 6

Program area Gilgit Chitral Baltistan Astore Total number

Area (square km) 28,500 14,850 25,850 5,000 74,200

Population (2000 est.) 383,542 318,689 313,826 70,000 1,086,057

Average household size 13 7 10 8 10

Number of rural households 29,050 42,492 31,072 8,850 111,464

Number of villages (approx.) 327 463 234 99 1,123

Potential VOs (no.) 654 956 699 199 2,508

Average population per VO 481 420 385 352 433

Average number of households per VO 36 56 38 44 44Source: AKRSP’s Monitoring, Evaluation, and Research Section. Population, household, and village data based on Northern Areas Census, 81/98.

P r o g r a m A r e a S o c i a l D a t a — G i l g i t , C h i t r a l ,B a l t i s t a n , a n d A s t o r e R e g i o n s , E n d - 2 0 0 0

T a b l e J . 2

A n n e x e s

7 7

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

Gilg

it

Num

ber o

f VOs

est

ablis

hed

129

286

312

345

371

409

446

470

494

501

539

559

676

701

768

810

828

798

Mem

bers

hip

(hou

seho

lds)

12,0

5023

,120

24,9

5026

,412

26,5

0026

,500

26,5

0026

,500

26,5

0026

,500

26,5

0024

,793

30,3

6631

,414

33,7

1835

,938

36,3

9135

,485

Perc

enta

ge o

f tot

al ru

ral h

ouse

hold

s32

6166

7070

7070

7070

7070

6580

8389

9596

94

37,9

00 in

VOs

in 2

000

Chitr

al

Num

ber o

f VOs

est

ablis

hed

4911

514

916

822

432

337

043

648

052

255

958

160

462

367

870

773

775

8

Mem

bers

hip

(hou

seho

lds)

3,39

97,

492

9,61

510

,667

12,3

8315

,236

16,6

5518

,574

19,8

5621

,124

21,7

4222

,601

23,3

2323

,778

25,6

7726

,884

27,8

5329

,116

Perc

enta

ge o

f tot

al ru

ral h

ouse

hold

s8

1823

2529

3639

4447

5051

5355

5660

6366

69

42,4

92 in

VOs

in 2

000

Bal

tista

n

Num

ber o

f VOs

est

ablis

hed

––

2253

159

247

342

446

514

565

597

630

675

741

771

782

806

815

Mem

bers

hip

(hou

seho

lds)

––

1,20

42,

882

6,14

59,

089

12,1

6217

,286

20,0

8422

,126

23,4

1624

,596

25,7

6428

,102

29,0

2229

,463

29,4

9829

,821

Perc

enta

ge o

f tot

al ru

ral h

ouse

hold

s–

–4

920

2939

5665

7175

7983

9093

9595

96

31,0

72 in

VOs

in 2

000

Tota

l

Num

ber o

f VOs

est

ablis

hed

178

401

483

566

754

979

1,15

81,

352

1,48

81,

588

1,69

51,

770

1,95

52,

065

2,21

72,

299

2,31

72,

371

Mem

bers

hip

(hou

seho

lds)

15,4

4930

,612

35,7

6939

,961

45,0

2850

,825

55,3

1762

,360

66,4

4069

,750

71,6

5871

,990

79,4

5383

,294

88,4

1792

,285

93,7

4294

,422

Perc

enta

ge o

f tot

al ru

ral h

ouse

hold

s14

2732

3640

4650

5660

6364

6571

7579

8384

85

11,1

464

in V

Os in

200

0Cu

mu

la

ti

ve

G

ro

wt

h

of

V

il

la

ge

O

rg

an

iz

at

io

ns

,

19

83

–2

00

0T

ab

le

J

.3

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

7 8

Year Organizations (no.) Membership (no.) Bank deposits (Rs. millions)

Village Organizations

1983 178 15,449 0.84

1984 223 15,163 5.16

1985 82 5,157 4.55

1986 83 4,192 5.35

1987 188 5,067 15.09

1988 225 5,797 15.52

1989 179 4,492 14.27

1990 194 7,043 14.46

1991 136 4,080 14.12

1992 100 3,310 20.12

1993 140 3,550 32.21

1994 106 1,947 34.95

1995 121 4,208 32.38

1996 110 3,839 39.47

1997 152 5,123 32.91

1998 82 3,870 36.71

1999 18 1,455 8.87

2000 54 680 - 7.34

Total 2,371 94,422 319.64

Women’s Organizations

1984 76 3,701 0.53

1985 27 1,183 0.76

1986 21 614 0.65

1987 37 1,446 1.38

1988 69 1,913 1.49

1989 57 1,851 2.81

1990 106 3,484 2.71

1991 89 2,527 3.24

1992 84 2,538 4.09

1993 103 3,451 5.30

1994 94 3,079 9.96

1995 91 3,244 14.71

1996 126 3,856 14.36

1997 161 4,752 13.22

1998 117 3,930 21.13

1999 71 3,625 8.07

2000 68 1,861 6.20

Total 1,397 47,055 110.61

S o c i a l O r g a n i z a t i o n , M e m b e r s h i p , a n dB a n k D e p o s i t s ( y e a r l y a d d i t i o n s )

T a b l e J . 4

A n n e x e s

7 9

Chitr

alIn

dica

tor

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

Orga

niza

tion

Num

ber o

f wom

en’s

orga

niza

tions

2049

6410

413

616

117

819

422

125

428

732

935

338

2

Savi

ngs

(cum

ulat

ive)

Tota

l (Rs

. mill

ion)

0.17

0.43

20.

921.

822.

373.

313.

965.

46.

878.

7211

.27

14.0

716

.01

17.8

2

Per g

roup

(Rs.

)8,

500

8,57

114

,375

17,5

0017

,426

20,5

5922

,247

27,8

3531

,086

34,3

3139

,268

42,7

6645

,354

46,6

49

Cred

it

Shor

t ter

m (R

s. m

illio

n)0

00

00

0.00

30.

350.

087

0.2

0.22

0.27

0.35

60.

356

0.35

6

Num

ber o

f org

aniza

tions

00

00

01

24

78

1015

1515

Med

ium

term

(Rs.

mill

ion)

00

00

00.

10.

10.

10.

10.

10.

10.

10.

10.

1

Num

ber o

f org

aniza

tions

00

00

01

11

11

11

11

WOC

P (R

s. m

illio

n)–

––

––

0.18

40.

541.

564

3.64

6.77

11.5

516

.286

16.2

8616

.286

Num

ber o

f org

aniza

tions

––

––

513

4394

149

228

300

300

300

Prod

uctio

n pa

ckag

es a

nd tr

aini

ng

(num

ber o

f gro

ups

parti

cipa

ting)

Vege

tabl

e de

mon

stra

tion

Plot

s/ye

ar4

848

050

610

317

99

8498

120

Nur

serie

s/ye

ar4

93

34

00

00

30

00

0

Hom

e-ba

sed

poul

try/y

ear (

WOs

)2

07

6411

70

00

06

3612

83

Tota

l pro

duct

ion

pack

ages

/yea

r10

1758

6717

16

103

1718

4596

106

123

Su

mm

ar

y

In

di

ca

to

rs

o

f

Pr

og

re

ss

i

n

In

st

it

ut

io

n

Bu

il

di

ng

a

nd

Te

ch

no

lo

gy

T

ra

ns

fe

r

fo

r

Wo

me

n’s

P

ro

gr

am

s

in

C

hi

tr

al

,B

al

ti

st

an

,

an

d

Gi

lg

it

Ta

bl

e

J.

5

(con

tinue

d on

follo

win

g pa

ge)

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

8 0

Bal

tista

nIn

dica

tor

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

Orga

niza

tion

Num

ber o

f wom

en’s

orga

niza

tions

––

1349

8110

412

414

216

222

528

429

631

635

3

Savi

ngs

(cum

ulat

ive)

Tota

l (Rs

. mill

ion)

––

0.02

0.08

0.18

0.45

0.60

0.88

1.31

2.50

3.47

4.03

4.48

5.25

Per g

roup

(Rs.

)–

–1,

538

1,63

32,

222

4,32

74,

839

6,19

78,

086

11,1

1112

,218

13,6

1514

,177

14,8

73

Cred

it

Shor

t ter

m (R

s. m

illio

n)0

00

00

0.00

50.

005

0.00

50.

260.

260.

260.

260.

260.

26

Num

ber o

f org

aniza

tions

00

00

02

22

2121

2121

2121

Med

ium

term

(Rs.

mill

ion)

00

00

00

00

00

00

00

Num

ber o

f org

aniza

tions

00

00

00

00

00

00

00

WOC

P (R

s. m

illio

n)–

––

––

––

0.05

0.17

1.17

2.16

1.83

3.56

3.98

Num

ber o

f org

aniza

tions

––

––

––

–1

420

3656

7812

0

Prod

uctio

n pa

ckag

es a

nd tr

aini

ng

(num

ber o

f gro

ups

parti

cipa

ting)

Vege

tabl

e de

mon

stra

tion

Plot

s/ye

ar4

120

3532

90

1024

174

00

00

Nur

serie

s/ye

ar0

00

01

10

380

44

00

0

Hom

e-ba

sed

poul

try/y

ear (

WOs

)12

1341

9029

510

1126

918

919

20

00

Tota

l pro

duct

ion

pack

ages

/yea

r16

2541

125

6261

059

293

367

196

00

0

Su

mm

ar

y

In

di

ca

to

rs

o

f

Pr

og

re

ss

i

n

In

st

it

ut

io

n

Bu

il

di

ng

a

nd

Te

ch

no

lo

gy

T

ra

ns

fe

r

fo

r

Wo

me

n’s

P

ro

gr

am

s

in

C

hi

tr

al

,B

al

ti

st

an

,

an

d

Gi

lg

it

(

co

nt

in

ue

d)

Ta

bl

e

J.

5

A n n e x e s

8 1

Gilg

itIn

dica

tor

Up

to 8

619

8719

8819

8919

9019

9119

9219

9319

9419

9519

9619

9719

9819

9920

00

Orga

niza

tion

Num

ber o

f wom

en’s

orga

niza

tions

129

146

186

215

245

270

306

349

372

471

501

570

633

660

662

Savi

ngs

(cum

ulat

ive)

Tota

l (Rs

. mill

ion)

1.97

3.18

4.42

6.71

8.46

11.0

513

.93

18.1

826

.11

39.4

550

.77

60.4

878

.24

83.9

287

.538

Per g

roup

(Rs.

)15

,271

21,7

8123

,763

31,2

0934

,531

40,9

2645

,523

52,0

9270

,188

83,7

5810

1,33

710

6,10

512

3,60

212

7,15

213

2,23

3

Cred

it

Shor

t ter

m (R

s. m

illio

n)0.

010.

492.

062.

93.

073.

43.

733.

934.

154.

244.

244.

334.

347

4.34

74.

347

Num

ber o

f

orga

niza

tions

1062

185

226

238

267

283

302

313

323

323

326

328

328

328

Med

ium

term

(Rs.

mill

ion)

00.

20.

621.

000.

971.

091.

271.

271.

271.

271.

271.

512.

047

2.04

72.

047

Num

ber o

f

orga

niza

tions

05

1223

2325

2626

2626

2628

3333

33

WOC

P (R

s. m

illio

n)–

––

0.6

2.04

1.55

5.50

410

.399

19.6

0119

.41

59.8

893

.28

116.

629

116.

629

116.

629

Num

ber o

f

orga

niza

tions

––

–2

2544

7211

818

930

943

658

970

070

070

0

Prod

uctio

n pa

ckag

es a

nd

train

ing

(num

ber o

f gro

ups

parti

cipa

ting)

Vege

tabl

e de

mon

stra

tion

Plot

s/ye

ar0

2332

410

1246

12

426

223

21

Nur

serie

s/ye

ar6

01

126

010

041

04

20

00

Hom

e-ba

sed

poul

try/y

ear

(WOs

)0

03

599

2128

3637

187

174

316

410

Tota

l pro

duct

ion

pack

ages

/yea

r6

2336

112

1533

8437

8019

120

455

96

11

Su

mm

ar

y

In

di

ca

to

rs

o

f

Pr

og

re

ss

i

n

In

st

it

ut

io

n

Bu

il

di

ng

a

nd

Te

ch

no

lo

gy

T

ra

ns

fe

r

fo

r

Wo

me

n’s

P

ro

gr

am

s

in

C

hi

tr

al

,B

al

ti

st

an

,

an

d

Gi

lg

it

Ta

bl

e

J.

5

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

8 2

A: GILGITCompleted

Region/type of project Initiated (no.) Number Percent

Irrigation 368 313 85

Feeder channels/pipe irrigation 313 266 85

Lift irrigation 16 11 69

Storage reservoir 30 29 97

Siphon irrigation 5 4 80

Sedimentation tank 4 3 75

Channelization of rivers – – –

Transport 131 107 82

Link roads 94 78 83

Bridge/link road 22 18 82

Pony tracks 15 11 73

Foot bridges – – 0

Others 65 57 88

Protective works 44 40 91

Boundary walls 2 2 100

Nursery 5 5 100

Hydel scheme 9 7 78

Flour mill – – –

Mud-flow control 2 2 100

Super passage 1 1 100

Water supply/delivery 2 0 0

WO Projects 74 38 51

Water delivery/supply 55 25 45

Vocational schools 1 1 100

Pipe irrigation 1 1 100

Barbed wire 8 5 63

Link road 1 1 100

Agriculture nursery 1 1 100

Pony track 2 0 0

Sanitation 4 4 100

Washing center – – –

Microhydel 1 0 0

Community centers – – –

Community bathrooms – – –

Total 638 515 81

P r o d u c t i v e P h y s i c a l I n f r a s t r u c t u r e P r o j e c t s ,b y R e g i o n a n d T y p e ( c u m u l a t i v e )

T a b l e J . 6

A n n e x e s

8 3

B: CHITRALCompleted

Region/type of project Initiated (no.) Number Percent

Irrigation 315 298 95

Feeder channels/pipe irrigation 279 264 95

Lift irrigation 2 2 100

Storage reservoir 20 20 100

Siphon irrigation 10 9 90

Sedimentation tank 1 1 100

Channelization of rivers 3 2 67

Transport 213 200 94

Link roads 184 175 95

Bridge/link road 25 21 84

Pony tracks 3 3 100

Foot bridges 1 1 100

Others 246 222 90

Protective works 71 65 92

Boundary walls – – –

Nursery – – –

Hydel scheme 166 153 92

Flour mill 1 1 100

Mud-flow control – – –

Super passage – – –

Water supply/delivery 8 3 38

WO Projects 5 0 0

Water delivery/supply 4 0 0

Vocational schools – – –

Pipe irrigation – – –

Barbed wire – – –

Link road – – –

Agriculture nursery – – –

Pony track – – –

Sanitation – – –

Washing center 1 0 0

Microhydel – – –

Community centers – – –

Community bathrooms – – –

Total 779 720 92

P r o d u c t i v e P h y s i c a l I n f r a s t r u c t u r e P r o j e c t s ,b y R e g i o n a n d T y p e ( c u m u l a t i v e )

T a b l e J . 6

(continued on following page)

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

8 4

C: BALTISTANCompleted

Region/type of project Initiated (no.) Number Percent

Irrigation 436 369 85

Feeder channels/pipe irrigation 388 323 83

Lift irrigation 14 13 93

Storage reservoir 30 29 97

Siphon irrigation 4 4 100

Sedimentation tank – – –

Channelization of rivers – – –

Transport 143 118 83

Link roads 96 72 75

Bridge/link road 10 9 90

Pony tracks 37 37 100

Foot bridges 0 0 –

Others 209 185 89

Protective works 89 81 91

Boundary walls 78 76 97

Nursery – – –

Hydel scheme 9 7 78

Flour mill – – –

Mud-flow control – – –

Super passage – – –

Water supply/delivery 33 21 64

WO Projects 14 10 71

Water delivery/supply – – –

Vocational schools – – –

Pipe irrigation – – –

Barbed wire – – –

Link road – – –

Agriculture nursery – – –

Pony track – – –

Sanitation 1 1 100

Washing center – – –

Microhydel – – –

Community centers 5 2 40

Community bathrooms 8 7 88

Total 802 682 85

P r o d u c t i v e P h y s i c a l I n f r a s t r u c t u r e P r o j e c t s ,b y R e g i o n a n d T y p e ( c u m u l a t i v e ) ( c o n t i n u e d )

T a b l e J . 6

A n n e x e s

8 5

D: PROGRAM AREACompleted

Region/type of project Initiated (no.) Number Percent

Irrigation 1,119 980 88

Feeder channels/pipe irrigation 980 853 87

Lift irrigation 32 26 81

Storage reservoir 80 78 98

Siphon irrigation 19 17 89

Sedimentation tank 5 4 80

Channelization of rivers 3 2 67

Transport 487 425 87

Link roads 374 325 87

Bridge/link road 57 48 84

Pony tracks 55 51 93

Foot bridges 1 1 100

Others 520 464 89

Protective works 204 186 91

Boundary walls 80 78 98

Nursery 5 5 100

Hydel scheme 184 167 91

Flour mill 1 1 100

Mud-flow control 2 2 100

Super passage 1 1 100

Water supply/delivery 43 24 56

WO Projects 93 48 52

Water delivery/supply 59 25 42

Vocational schools 1 1 100

Pipe irrigation 1 1 100

Barbed wire 8 5 63

Link road 1 1 100

Agriculture nursery 1 1 100

Pony track 2 0 0

Sanitation 5 5 100

Washing center 1 0 0

Microhydel 1 0 0

Community centers 5 2 40

Community bathrooms 8 7 88

Total 2,219 1,917 86

P r o d u c t i v e P h y s i c a l I n f r a s t r u c t u r e P r o j e c t s ,b y R e g i o n a n d T y p e ( c u m u l a t i v e )

T a b l e J . 6

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

8 6

A: GILGITNo of AKRSP grant Disbursement Completion

Region/type of project projects (Rs.’000) Rs.’000 Percent Number Percent

Irrigation 368 122,715 92,955 76 313 85

Feeder channels 294 97,617 69,544 71 248 84

Pipe/siphon irrigation 23 8,247 7,584 92 21 91

Lift irrigation 16 10,502 9,552 91 11 69

Storage reservoirs 30 5,521 5,546 100 29 97

Sedimentation tank 4 639 540 85 3 75

Tunnel (irrigation channel) 1 189 189 100 1 100

Channelization of rivers – – – – – –

Transport 131 43,033 31,925 74 107 82

Link roads 94 29,152 20,667 71 78 83

Bridge/link roads 22 9,147 6,728 74 18 82

Pony tracks 15 4,734 4,530 96 11 73

Foot bridges – – – – – –

Others 65 22,898 16,339 71 57 88

Protective works 44 11,244 10,100 90 40 91

Boundary walls 2 192 184 96 2 100

Nursery 5 489 450 92 5 100

Hydel scheme 9 9,617 4,985 52 7 78

Flour mill 0 0 0 0 0 0

Mud-flow control 2 220 202 92 2 100

Super passage 1 189 161 85 1 100

Water supply/delivery 2 947 257 27 0 0

WO Projects 74 29,102 12,716 44 38 51

Water delivery/supply 55 25,545 11,194 44 25 45

Vocational schools 1 50 50 100 1 100

Pipe irrigation 1 304 304 100 1 100

Barbed wire 8 1,646 491 30 5 63

Link road 1 92 92 100 1 100

Agriculture nursery 1 50 49 98 1 100

Pony track 2 570 318 56 – 0

Sanitation/washing center 4 218 218 100 4 100

Microhydel 1 627 – 0 – 0

Community centers 0 0 0 0 0 0

Community bathrooms 0 0 0 0 0 0

Total 638 217,748 153,935 71 515 81

P r o d u c t i v e I n f r a s t r u c t u r e P r o j e c t s , C o s t s , a n dC o n s t r u c t i o n P r o g r e s s , b y T y p e o f P r o j e c t , a so f D e c e m b e r 2 0 0 0 ( c u m u l a t i v e )

T a b l e J . 7

A n n e x e s

8 7

B: CHITRALNo of AKRSP grant Disbursement Completion

Region/type of project projects (Rs.’000) Rs.’000 Percent Number Percent

Irrigation 315 90,379 77,152 85 298 95

Feeder channels 266 68,677 63,170 92 255 96

Pipe/siphon irrigation 23 15,670 8,201 52 18 78

Lift irrigation 2 351 333 95 2 100

Storage reservoirs 20 4,454 4,392 99 20 100

Sedimentation tank 1 309 309 100 1 100

Tunnel (irrigation channel) 0 0 0 0 0 0

Channelization of rivers 3 918 747 81 2 67

Transport 213 67,534 59,985 89 200 94

Link roads 184 55,105 48,889 89 175 95

Bridge/link roads 25 11,372 10,043 88 21 84

Pony tracks 3 752 748 99 3 100

Foot bridges 1 305 305 100 1 100

Others 246 143,130 129,504 90 222 90

Protective works 71 20,328 18,559 91 65 92

Boundary walls 0 0 0 0 0 0

Nursery 0 0 0 0 0 0

Hydel scheme 166 114,282 107,182 94 153 92

Flour mill 1 141 141 100 1 100

Mud-flow control 0 0 0 0 0 0

Super passage 0 0 0 0 0 0

Water supply/delivery 8 8,379 3,622 43 3 38

WO Projects 5 3,410 2,377 70 - 0

Water delivery/supply 4 3,016 2,063 68 0

Vocational schools 0 0 0 0 0 0

Pipe irrigation 0 0 0 0 0 0

Barbed wire 0 0 0 0 0 0

Link road 0 0 0 0 0 0

Agriculture nursery 0 0 0 0 0 0

Pony track 0 0 0 0 0 0

Sanitation/washing center 1 394 314 80 - 0

Microhydel 0 0 0 0 0 0

Community centers 0 0 0 0 0 0

Community bathrooms 0 0 0 0 0 0

Total 779 304,453 269,018 88 720 92

P r o d u c t i v e I n f r a s t r u c t u r e P r o j e c t s , C o s t s , a n dC o n s t r u c t i o n P r o g r e s s , b y T y p e o f P r o j e c t , a so f D e c e m b e r 2 0 0 0 ( c u m u l a t i v e )

T a b l e J . 7

(continued on following page)

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

8 8

C: BALTISTANNo of AKRSP grant Disbursement Completion

Region/type of project projects (Rs.’000) Rs.’000 Percent Number Percent

Irrigation 436 132,616 93,459 70 369 85

Feeder channels 345 103,732 73,217 71 283 82

Pipe/siphon irrigation 47 21,571 14,016 65 44 94

Lift irrigation 14 3,930 3,005 76 13 93

Storage reservoirs 30 3,383 3,221 95 29 97

Sedimentation tank 0 0 0 0 0 0

Tunnel (irrigation channel) 0 0 0 0 0 0

Channelization of rivers 0 0 0 0 0 0

Transport 143 51,769 41,107 79 118 83

Link roads 96 42,930 33,652 78 72 75

Bridge/link roads 10 4,642 3,351 72 9 90

Pony tracks 37 4,197 4,104 98 37 100

Foot bridges 0 0 0 0 0 0

Others 209 59,704 50,762 85 185 89

Protective works 89 28,996 23,371 81 81 91

Boundary walls 78 9,655 9,324 97 76 97

Nursery 0 0 0 0 0 0

Hydel scheme 9 6,454 6,213 96 7 78

Flour mill 0 0 0 0 0 0

Mud-flow control 0 0 0 0 0 0

Super passage 0 0 0 0 0 0

Water supply/delivery 33 14,599 11,854 81 21 64

WO Projects 14 1,649 1,162 70 10 71

Water delivery/supply 0 0 0 0 0 0

Vocational schools 0 0 0 0 0 0

Pipe irrigation 0 0 0 0 0 0

Barbed wire 0 0 0 0 0 0

Link road 0 0 0 0 0 0

Agriculture nursery 0 0 0 0 0 0

Pony track 0 0 0 0 0 0

Sanitation/washing center 1 56 56 100 1 100

Microhydel 0 0 0 0 0 0

Community centers 5 1,093 630 58 2 40

Community bathrooms 8 500 476 95 7 88

Total 802 245,738 186,490 76 682 85

P r o d u c t i v e I n f r a s t r u c t u r e P r o j e c t s , C o s t s , a n dC o n s t r u c t i o n P r o g r e s s , b y T y p e o f P r o j e c t , a so f D e c e m b e r 2 0 0 0 ( c u m u l a t i v e ) ( c o n t i n u e d )

T a b l e J . 7

A n n e x e s

8 9

D: PROGRAM AREANo of AKRSP grant Disbursement Completion

Region/type of project projects (Rs.’000) Rs.’000 Percent Number Percent

Irrigation 1,119 345,710 263,566 76 980 88

Feeder channels 905 270,026 205,931 76 786 87

Pipe/siphon irrigation 93 45,488 29,801 66 83 89

Lift irrigation 32 14,783 12,890 87 26 81

Storage reservoirs 80 13,358 13,159 99 78 98

Sedimentation tank 5 948 849 90 4 80

Tunnel (irrigation channel) 1 189 189 100 1 100

Channelization of rivers 3 918 747 81 2 67

Transport 487 162,336 133,017 82 425 87

Link roads 374 127,187 103,208 81 325 87

Bridge/link roads 57 25,161 20,122 80 48 84

Pony tracks 55 9,683 9,382 97 51 93

Foot bridges 1 305 305 100 1 100

Others 520 225,732 196,605 87 464 89

Protective works 204 60,568 52,030 86 186 91

Boundary walls 80 9,847 9,508 97 78 98

Nursery 5 489 450 92 5 100

Hydel scheme 184 130,353 118,380 91 167 91

Flour mill 1 141 141 100 1 100

Mud-flow control 2 220 202 92 2 100

Super passage 1 189 161 85 1 100

Water supply/delivery 43 23,925 15,733 66 24 56

WO Projects 93 34,161 16,255 48 48 52

Water delivery/supply 59 28,561 13,257 46 25 42

Vocational schools 1 50 50 100 1 100

Pipe irrigation 1 304 304 100 1 100

Barbed wire 8 1,646 491 30 5 63

Link road 1 92 92 100 1 100

Agriculture nursery 1 50 49 98 1 100

Pony track 2 570 318 56 - 0

Sanitation/washing center 6 668 588 88 5 83

Microhydel 1 627 - 0 - 0

Community centers 5 1,093 630 58 2 40

Community bathrooms 8 500 476 95 7 88

Total 2,219 767,939 609,443 79 1,917 86

P r o d u c t i v e I n f r a s t r u c t u r e P r o j e c t s , C o s t s , a n dC o n s t r u c t i o n P r o g r e s s , b y T y p e o f P r o j e c t , a so f D e c e m b e r 2 0 0 0 ( c u m u l a t i v e )

T a b l e J . 7

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

9 0

Gilgit Chitral Baltistan Total

Number of projects initiated 638 779 802 2,219

Cost of projects initiated (Rs. millions) 217.75 304.45 245.74 767.94

Number of beneficiary households 64,996 49,215 52,524 166,735

Physical progress of projects 81% 90% 85% 84%

Number of projects completed 515 720 682 1,917

S u m m a r y D a t a o f P r o d u c t i v e P h y s i c a lI n f r a s t r u c t u r e P r o j e c t s b y R e g i o n ,D e c e m b e r 2 0 0 0

T a b l e J . 8

PPIs initiated Cost Completed PPIs Year (no.) (Rs. millions) per year (no.)

1983 104 15.30 23

1984 112 17.50 91

1985 103 21.90 81

1986 93 17.50 62

1987 159 24.60 116

1988 186 32.30 139

1989 144 25.49 118

1990 166 35.10 133

1991 137 26.71 119

1992 111 23.04 113

1993 110 26.07 128

1994 76 19.49 81

1995 122 44.45 109

1996 94 41.17 119

1997 97 67.83 109

1998 149 102.78 139

1999 137 84.79 83

2000 119 239.94 154

Total 2,219 865.96 1,917Note: Year additions may not tally due to ongoing revisions.

P r o d u c t i v e P h y s i c a lI n f r a s t r u c t u r e( y e a r l y a d d i t i o n s )

T a b l e J . 9

A n n e x e s

9 1

App

rop.

Farm

Cl

uste

r Li

tera

cy/

Voca

tiona

l A

ll Ye

arA

gric

ultu

reFo

rest

ryLi

vest

ock

Poul

try

Mar

ketin

gA

ccou

nts

tech

nolo

gym

achi

nery

hatc

hery

TBA

snu

mer

acy

trai

ning

subj

ects

Regu

lar C

ours

es

1983

460

450

00

00

00

00

91

1984

450

6169

00

00

00

00

175

1985

670

8572

380

00

00

00

262

1986

136

089

5849

00

00

00

033

2

1987

143

012

458

4916

00

00

00

390

1988

207

020

021

748

3328

00

00

073

3

1989

331

5012

413

947

6074

00

00

082

5

1990

481

4815

829

047

153

370

00

00

1,21

4

1991

339

138

111

156

2130

022

231

00

00

1,31

8

1992

254

196

9116

218

274

00

40

00

999

1993

245

104

124

969

216

320

80

00

834

1994

205

9081

184

6490

590

80

00

781

1995

200

173

1224

517

236

830

00

00

01,

200

1996

330

216

210

208

235

285

400

06

60

1,53

6

1997

124

121

481

144

622

417

600

00

5117

72,

197

1998

306

131

120

3844

540

00

010

2315

21,

265

1999

168

438

9286

1,18

130

00

011

345

341

2,49

4

2000

228

1197

675

366

610

00

3412

663

62,

234

Tota

l3,

855

1,71

62,

305

2,89

73,

411

2,34

358

231

2016

325

11,

306

18,8

80

Th

e

Tr

ai

ni

ng

P

ro

gr

am

—N

um

be

r

of

T

ra

in

ee

sT

ab

le

J

.1

0

(con

tinue

d on

follo

win

g pa

ge)

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

9 2

App

rop.

Farm

Cl

uste

r Li

tera

cy/

Voca

tiona

l A

ll Ye

arA

gric

ultu

reFo

rest

ryLi

vest

ock

Poul

try

Mar

ketin

gA

ccou

nts

tech

nolo

gym

achi

nery

hatc

hery

TBA

snu

mer

acy

trai

ning

subj

ects

Refr

eshe

r and

Oth

er C

ours

es

1984

890

600

00

00

00

00

149

1985

270

6814

00

00

00

00

109

1986

970

151

1832

00

00

00

029

8

1987

119

014

125

240

00

00

00

309

1988

129

076

1515

220

00

00

025

7

1989

700

195

1223

00

00

00

030

0

1990

369

2547

589

210

00

00

00

979

1991

390

4353

660

200

370

00

00

1,08

6

1992

440

6863

517

823

115

00

00

00

1,45

9

1993

1,06

968

798

325

2518

80

00

00

02,

473

1994

988

124

876

400

161

339

60

00

00

2,89

4

1995

570

128

500

243

2215

20

00

00

01,

615

1996

506

223

765

344

029

90

00

60

02,

143

1997

423

5539

832

30

325

00

00

00

1,52

4

1998

396

8131

318

511

828

90

00

1017

191

1,60

0

1999

234

106

597

302

6113

70

00

2327

163

1,65

0

2000

102

1641

017

710

927

80

00

1948

583

1,74

2

Tota

l6,

018

937

6,99

42,

710

654

2,14

443

00

5892

937

20,5

87N

ote:

Annu

al to

tals

may

not

tally

due

to o

ngoi

ng re

visi

ons.

The

Cor

e Of

fice

has

not m

aint

aine

d th

e an

nual

bre

akdo

wn

sinc

e 19

98 d

ue to

a c

hang

e in

M&

E re

porti

ng s

yste

ms.

Th

e

Tr

ai

ni

ng

P

ro

gr

am

—N

um

be

r

of

T

ra

in

ee

s

(c

on

ti

nu

ed

)T

ab

le

J

.1

0

A n n e x e s

9 3

App

rop.

Farm

Cl

uste

r Li

tera

cy/

Voca

tiona

l A

ll Ye

arA

gric

ultu

reFo

rest

ryLi

vest

ock

Poul

try

Mar

ketin

gA

ccou

nts

tech

nolo

gym

achi

nery

hatc

hery

TBA

snu

mer

acy

trai

ning

subj

ects

Regu

lar C

ours

es

1983

20

20

00

00

00

00

4

1984

20

33

00

00

00

00

8

1985

50

43

20

00

00

00

14

1986

80

51

20

00

00

00

16

1987

90

54

41

00

00

00

23

1988

120

910

42

20

00

00

39

1989

82

610

43

50

00

00

38

1990

253

715

412

50

00

00

71

1991

217

613

224

211

00

00

95

1992

1510

515

221

00

10

00

69

1993

187

78

114

30

10

00

59

1994

165

514

55

40

10

00

55

1995

1613

218

1221

20

00

00

84

1996

2219

1415

1318

30

01

10

106

1997

1514

3414

8130

10

00

26

197

1998

2212

92

273

00

01

213

91

1999

1643

128

118

20

00

63

1222

0

2000

151

556

274

00

02

529

144

Tota

l24

713

614

020

930

816

046

13

1013

601,

333

Th

e

Tr

ai

ni

ng

P

ro

gr

am

—N

um

be

r

of

C

ou

rs

es

Ta

bl

e

J.

11

(con

tinue

d on

follo

win

g pa

ge)

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

9 4

App

rop.

Farm

Cl

uste

r Li

tera

cy/

Voca

tiona

l A

ll Ye

arA

gric

ultu

reFo

rest

ryLi

vest

ock

Poul

try

Mar

ketin

gA

ccou

nts

tech

nolo

gym

achi

nery

hatc

hery

TBA

snu

mer

acy

trai

ning

subj

ects

Refr

eshe

r and

Oth

er C

ours

es

1984

20

30

00

00

00

00

5

1985

30

51

00

00

00

00

9

1986

60

61

20

00

00

00

15

1987

70

32

20

00

00

00

14

1988

40

31

21

00

00

00

11

1989

30

51

20

00

00

00

11

1990

192

206

20

00

00

00

49

1991

223

325

30

100

00

00

75

1992

234

268

28

00

00

00

71

1993

394

3315

28

00

00

00

101

1994

369

3329

818

10

00

00

134

1995

3310

1922

19

00

00

00

94

1996

257

3119

020

00

01

00

103

1997

2920

2118

017

00

00

00

105

1998

306

1612

915

00

01

12

92

1999

267

3019

711

00

01

12

104

2000

101

2511

1121

00

01

218

100

Tota

l31

773

311

170

5312

811

00

44

221,

093

Not

e:An

nual

tota

ls m

ay n

ot ta

lly d

ue to

ong

oing

revi

sion

s. T

he C

ore

Offic

e ha

s no

t mai

ntai

ned

the

annu

al b

reak

dow

n si

nce

1998

due

to a

cha

nge

in M

&E

repo

rting

sys

tem

s.

Th

e

Tr

ai

ni

ng

P

ro

gr

am

—N

um

be

r

of

C

ou

rs

es

(

co

nt

in

ue

d)

Ta

bl

e

J.

11

Not

e:Th

is ta

ble

has

not b

een

upda

ted

sinc

e 19

97 b

ecau

se th

e ne

w L

F-ba

sed

M&

E sy

stem

s/pr

ogre

ss re

ports

for t

he c

urre

nt p

hase

do

not a

ccom

mod

ate

it. T

he ta

ble

is re

tain

ed o

nly

to k

eep

num

berin

g co

nsis

tent

with

pre

viou

s OE

D ev

alua

tions

.

Info

rmat

ion

on th

e ab

ove

indi

cato

rs is

ava

ilabl

e in

regi

onal

offi

ces.

AK

RS

P

Pr

od

uc

ti

on

P

ac

ka

ge

s

fo

r

Wo

me

n’s

O

rg

an

iz

at

io

ns

,

by

Re

gi

on

,

Cu

mu

la

ti

ve

t

o D

ec

em

be

r

31

,

20

00

Ta

bl

e

J.

12

A n n e x e s

9 5

School levelPrimary Middle Secondary Total

Gilgit

Government

Male 237 66 48 351

Female 136 15 14 165

Both 251 5 2 258

AKES

Male 2 3 1 6

Female 3 5 7 15

Both 65 28 13 106

Chitral

Government

Male 463 43 42 548

Female 153 24 5 182

Both 0 0 0 0

AKES

Male 10 0 0 10

Female 51 16 8 75

Both 0 0 0 0

Baltistan

Government

Male 175 40 31 246

Female 89 15 6 110

Both 224 12 4 240

Total

Government

Male 875 149 121 1,145

Female 378 54 25 457

Both 475 17 6 498

AKES

Male 12 3 1 16

Female 54 21 15 90

Both 65 28 13 106

Girls’ schools (%) 49 49 33 47

G i r l s ’ a n d B o y s ’ S c h o o l s i n t h e N o r t h e r nA r e a s a n d C h i t r a l , 1 9 9 4

T a b l e J . 1 3

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

9 6

Per-

1982

–ce

nt-

Don

or83

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

Tota

lag

eAg

a Kh

an F

ound

atio

n (A

KF)

Net

wor

k14

,852

10,0

4812

,819

17,4

1114

,200

8,34

69,

253

8,26

218

,481

8,42

27,

079

15,3

3911

,121

5,98

94,

082

2,62

55,

968

6,44

518

0,74

27

AKF

Paki

stan

10,9

607,

160

5,90

17,

663

7,77

85,

251

7,81

96,

959

17,0

787,

101

5,08

612

,676

11,1

215,

989

4,08

22,

625

5,96

86,

445

137,

662

5

AKF

Cana

da1,

420

2,35

71,

398

1,19

92,

013

1,10

772

01,

303

1,40

31,

321

1,99

32,

663

18,8

971

AKF

USA

1,32

93,

193

4,43

03,

992

12,9

440

AKF

UK1,

143

531

2,32

74,

119

417

1,98

871

411

,239

0

Cana

dian

Inte

rnat

iona

l

Deve

lopm

ent A

genc

y5,

439

9,24

511

,240

14,7

7818

,595

27,5

3338

,093

15,0

6220

,635

46,0

4437

,401

77,0

6325

,837

93,2

3625

,963

23,8

8826

,238

33,0

2554

9,31

520

Albe

rta A

id23

62,

192

3,48

64,

062

467

545

10,9

880

US A

genc

y fo

r Int

erna

tiona

l

Deve

lopm

ent

1,42

64,

097

5,82

05,

950

17,2

931

Ford

Fou

ndat

ion

1,32

549

61,

821

0

Heife

r Int

erna

tiona

l1,

518

4013

91,

697

0

Over

seas

Dev

elop

men

t

Agen

cy32

791

32,

092

3,47

118

133

,809

21,9

4622

,013

65,9

0017

,977

34,5

8219

,316

36,7

4023

,446

30,9

2728

9,46

120

4,53

314

1,47

494

9,10

834

OXFA

M60

355

254

41,

447

1,70

128

353

082

285

77,

339

0

Com

mon

Eur

opea

n Co

mm

unity

2,39

12,

428

3,48

19,

069

4,61

816

,046

13,7

262,

281

39,6

6953

,076

39,2

7732

,665

218,

727

8

East

ern

Elec

trici

ty17

688

249

513

0

The

Net

herla

nds

Gove

rnm

ent

4,10

97,

669

20,5

8430

,899

37,6

6215

,386

26,6

6037

,340

31,1

7741

,324

42,3

5113

5,46

843

0,62

916

Konr

ad A

dena

uer F

ound

atio

n1,

299

1,81

43,

387

10,7

314,

595

4,02

610

,525

1,82

21,

199

39,3

981

Wor

ld C

onse

rvat

ion

Unio

n1,

198

1,70

12,

899

0

Nor

weg

ian

Agen

cy fo

r

Inte

rnat

iona

l Dev

elop

men

t14

,034

13,4

9815

,392

22,5

8620

,265

16,1

8723

,177

24,7

8828

,728

29,3

4820

8,00

38

Deut

sche

Ges

ells

chaf

t fur

Tech

nisc

he Z

usam

men

arbe

it/

BMZ

8,89

74,

906

7,34

916

,442

15,8

0822

,549

16,6

406,

086

98,6

774

The

Wor

ld B

ank

860

2,61

03,

470

0

Loca

l bod

ies

and

Rura

l

Deve

lopm

ent D

epar

tmen

t89

8930

208

0

Nor

ther

n Ar

eas

Coun

cil

884

195

1,07

90

Gove

rnm

ent o

f Pak

ista

n3,

048

334

3,38

20

JICA

1,30

994

175

93,

009

0

SDC

990

14,0

009,

270

24,2

601

Tota

l22

,782

23,0

3936

,301

51,3

0570

,777

110,

587

109,

771

69,2

5616

8,03

313

3,52

715

6,09

320

8,95

314

7,76

615

7,44

027

5,09

441

7,37

733

6,14

325

8,31

32,

752,

557

100

US$’

000

equi

vale

nt1,

739

1,57

22,

279

3,08

14,

068

6,14

45,

344

3,19

27,

060

5,32

45,

555

6,83

54,

692

4,38

46,

722

9,28

76,

843

4,89

189

,012

AK

RS

P

So

ur

ce

s

of

I

nc

om

e,

b

y

Do

no

r (

ru

pe

es

‘

00

0s

)T

ab

le

J

.1

4

A n n e x e s

9 7

Per-

1982

–ce

nt-

App

licat

ion/

indi

cato

r83

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

Tota

lag

eEx

pend

iture

Capi

tal e

xpen

ditu

re1,

931

1,99

22,

673

3,71

53,

367

7,19

13,

203

5,00

211

,315

8,71

515

,797

2,23

75,

678

14,8

049,

372

15,6

3772

,625

6,04

419

1,29

87

Sala

ries

& c

onsu

ltanc

ies

1,79

82,

577

3,89

36,

305

9,81

218

,907

20,9

6125

,557

31,8

3134

,774

36,7

7460

,151

74,8

3310

4,79

981

,004

93,6

0287

,337

96,2

7579

1,19

029

Trav

el48

547

174

21,

161

1,42

04,

251

4,13

05,

620

7,02

26,

726

7,42

57,

404

9,99

19,

542

10,5

0511

,416

12,9

7613

,339

114,

626

4

Adm

inis

tratio

n36

131

141

239

196

5,05

94,

636

5,54

53,

421

6,51

69,

249

5,22

26,

982

8,50

88,

736

10,3

2411

,062

9,47

895

,481

3

Offic

e m

aint

enan

ce37

057

61,

076

1,82

82,

279

1,67

12,

311

2,90

76,

753

4,77

25,

307

4,44

68,

142

6,91

27,

285

10,9

829,

206

13,0

9689

,919

3

Vehi

cle

oper

atio

n26

434

157

751

552

72,

750

3,53

74,

688

7,09

87,

333

9,59

210

,382

14,4

6114

,045

15,0

1817

,443

15,1

6714

,865

138,

603

5

WB

eval

uatio

n6,

213

6,21

30

Tota

l ope

ratin

g co

st4,

884

6,08

89,

102

13,7

6317

,601

39,8

2938

,778

49,3

1967

,440

68,8

3684

,144

89,8

4212

6,30

015

8,61

013

1,92

015

9,40

420

8,37

315

3,09

71,

427,

330

52

Trai

ning

296

913

2,75

24,

575

6,83

94,

794

8,96

08,

944

8,89

510

,122

11,1

9711

,221

15,3

9117

,596

18,0

6620

,317

20,8

6219

,631

191,

371

7

Rese

arch

, sur

vey,

and

dem

onst

ratio

n1,

726

3,28

94,

492

3,80

29,

414

3,05

43,

916

9,66

311

,688

13,2

4816

,859

22,5

9530

,104

31,4

9339

,077

45,7

4657

,986

59,5

3136

7,68

313

Infra

stru

ctur

e pr

ojec

ts7,

204

12,0

8313

,665

15,4

9020

,519

26,3

9727

,046

35,9

2827

,411

24,5

7526

,879

19,3

2631

,781

33,0

2559

,357

79,8

4684

,294

88,7

4163

3,56

723

Cred

it pr

ogra

m (n

et)

517

153

5,74

56,

727

15,0

6513

,796

8,87

42,

290

707

12,8

7429

,584

31,6

4336

,400

43,4

5966

,202

(21,

446)

(51,

277)

(51,

553)

149,

760

5

Tota

l non

oper

atin

g co

st9,

743

16,4

3826

,654

30,5

9451

,837

48,0

4148

,796

56,8

2548

,701

60,8

1984

,519

84,7

8511

3,67

612

5,57

318

2,70

212

4,46

311

1,86

511

6,35

01,

342,

381

48

Gra

nd to

tal

14,6

2722

,526

35,7

5644

,357

69,4

3887

,870

87,5

7410

6,14

411

6,14

112

9,65

516

8,66

317

4,62

723

9,97

628

4,18

331

4,62

228

3,86

732

0,23

826

9,44

72,

769,

711

100

US$’

000

equi

vale

nt1,

116

1,60

32,

244

2,66

43,

990

4,88

14,

263

4,89

14,

882

5,16

96,

002

5,71

37,

621

7,91

47,

689

6,31

76,

520

4,21

083

,477

Fact

or to

brin

g to

199

9 ru

pees

3.81

3.48

3.32

3.22

3.08

2.81

2.58

2.42

2.15

1.96

1.80

1.59

1.39

1.29

1.14

1.06

1.00

0.94

Exch

ange

rate

13.1

114

.05

15.9

316

.65

17.4

018

.00

20.5

421

.70

23.7

925

.08

28.1

030

.57

31.4

935

.91

40.9

244

.94

49.1

264

.00

Op. c

ost (

’000

) in

1999

rupe

es18

,602

21,1

6330

,202

44,3

2554

,290

111,

828

99,9

1011

9,34

714

4,79

813

4,59

915

1,04

514

2,57

517

6,17

620

4,32

715

0,08

016

8,23

820

8,37

314

3,91

1

Effic

ienc

y in

dica

tors

Adm

in a

s %

of t

otal

01

01

06

55

35

53

33

34

34

Op. c

ost a

s %

of t

otal

exc

l. cr

edit

3224

2427

2236

3438

4845

4538

3737

2530

3934

52

Sala

ries/

cons

as

% o

f tot

al12

1111

1414

2224

2427

2722

3431

3726

3327

3629

Trav

el a

s %

of t

otal

32

23

25

55

65

44

43

34

45

4

Trai

ning

as

% o

f tot

al2

48

1010

510

88

87

66

66

77

77

Res.

/sur

v./de

mo

as %

of t

otal

1215

139

143

49

1010

1013

1311

1216

1822

13

Infra

stru

ctur

e as

% o

f tot

al49

5438

3530

3031

3424

1916

1113

1219

2826

3323

Cred

it ne

t as

% o

f tot

al4

116

1522

1610

21

1018

1815

1521

-8-1

6-1

95

Tota

l no.

of V

Os17

840

148

356

675

497

91,

158

1,35

21,

488

1,58

81,

695

1,77

01,

955

2,06

52,

217

2,29

92,

317

2,37

1

Tota

l no.

of h

ouse

hold

s15

,449

30,6

1235

,769

39,9

6145

,028

50,8

2555

,317

62,3

6066

,440

69,7

5071

,658

71,9

9079

,453

83,2

9488

,417

92,2

8593

,742

94,4

22

Hous

ehol

ds p

er c

omm

unity

8776

7471

6052

4846

4544

4241

4140

4040

4040

$ Op

. cos

t/hh

in 1

999

$25

1417

2325

4537

3944

3943

4045

5035

3745

45

$ In

frast

. cos

t/hh

in 1

999

$36

2826

2529

3026

2818

1414

911

1016

1918

15

$ Cr

edit

per h

ouse

hold

30

1111

2116

82

07

1514

1314

17-5

-11

-9So

urce

:AKR

SP a

nd O

ED a

naly

sis.

AK

RS

P:

A

pp

li

ca

ti

on

o

f

Fu

nd

s

Si

nc

e

St

ar

t-

Up

a

nd

E

ff

ic

ie

nc

yI

nd

ic

at

or

s

(r

up

ee

s

’0

00

s)

Ta

bl

e

J.

15

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

9 8

Dis

tri-

butio

n Ca

tego

ry19

8719

8819

8919

9019

9119

9219

9319

9419

9519

9619

9719

9819

9920

0020

01(%

)

Man

agem

ent

910

1013

1313

1213

811

1112

119

92

Seni

or p

rofe

ssio

nals

2116

2550

4952

5553

3622

3429

4545

4713

Juni

or p

rofe

ssio

nals

6162

7810

011

210

510

412

615

216

316

616

215

715

216

345

Supp

ort

5349

5559

5157

5677

3534

3633

2218

175

Driv

ers/

auxi

liarie

s89

100

102

9510

011

612

717

211

712

112

513

612

712

712

835

Tota

l23

323

727

031

732

534

335

444

134

835

137

237

236

235

136

410

0

Perc

ent i

ncre

ase

214

173

63

25-2

11

60

-3-3

4

AK

RS

P

St

af

f

St

re

ng

th

,

19

87

–2

00

0T

ab

le

J

.1

6

AKR

SPN

RSP

SRSC

PRSP

GB

TILP

RPSG

ATR

DP

Tota

lCa

tego

ry19

8319

9219

8919

9819

9519

9719

7219

98

Dist

ricts

68

418

Villa

ge, c

omm

unity

, and

wom

en’s

orga

niza

tions

3,76

811

,661

3,45

03,

290

139

312

214

177

23,0

11

Mem

bers

hip

141,

477

259,

490

104,

747

69,3

192,

695

10,1

601,

082

14,7

6060

3,73

0

Aver

age

mem

bers

hip

of o

rgan

izatio

ns38

2230

2119

335

8326

VO/C

O/W

O sa

ving

s (R

s. m

illio

ns)

430.

2529

5.26

43.9

348

.75

0.38

3.06

-5.

6482

7.27

Num

ber o

f vill

age

spec

ialis

ts tr

aine

d22

,681

68,2

3140

,890

31,7

011,

361

2,35

325

4,09

717

1,33

9

Amou

nt o

f cre

dit d

isbu

rsed

(Rs.

mill

ions

)1,

450.

402,

285.

9816

2.32

610.

3912

.60

6.10

0.21

19.4

44,

547.

43

Num

ber o

f cre

dit b

enef

icia

ries

590,

100

145,

484

16,5

0544

,984

875

345

802,

694

801,

067

Num

ber o

f phy

sica

l inf

rast

ruct

ure

proj

ects

com

plet

ed1,

917

2,31

11,

424

3062

7311

2,38

08,

208

Co

ns

ol

id

at

ed

P

ic

tu

re

o

f

Ru

ra

l

Su

pp

or

t

Pr

og

ra

ms

,

as

o

fD

ec

em

be

r

20

00

(

nu

mb

er

,

un

le

ss

o

th

er

wi

se

n

ot

ed

)T

ab

le

J

.1

7

9 9

ANNEX K: ORGANIZATIONAL CHARTS

Ag

a

Kh

an

R

ur

al

S

up

po

rt

P

ro

gr

am

C

or

e

an

d

Re

gi

on

al

O

ff

ic

es

Fi

gu

re

K

.1

Boar

d of

Dire

ctor

s

Gene

ral

Man

ager

Exec

utiv

eSe

cret

ary

RPO–

Balti

stan

RPM

Mic

ro-F

inan

ceEn

terp

rise

Deve

lopm

ent

Polic

y &

Rese

arch

Fina

nce

HR &

ADM

INRP

O–Gi

lgit

RPM

RPO–

Chitr

al

Isla

mab

adPM

–P &

RIs

lam

abad

Dire

ctor

of F

inan

ce

Man

ager

HR

HR–M

onito

r

Acco

unta

nt -

Core

Isla

mab

adPM

–HR

& A

-Co

re

Isla

mab

adPM

–Fin

ance

-Co

re

Coor

d.Li

nkag

esPM

–C &

S

PM–E

D

B. D

ev.

Coun

cilo

r

M&

ECo

nsul

tant

Coor

dina

tor

Syst

emAn

alys

t

Libr

aria

n

Mon

itor

Econ

omis

t

Fina

ncia

l An

alys

t

Loan

Por

tfolio

Audi

tor

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

1 0 0

Ag

a

Kh

an

R

ur

al

S

up

po

rt

P

ro

gr

am

(

RP

O,

e

.g

.,

B

al

ti

st

an

)F

ig

ur

e

K.

2

GaD

Coor

dSe

cret

ary

Khap

luM

anag

erM

anag

erM

icro

finan

ce

Man

ager

Mon

itorin

gEv

alua

tion

& R

esea

rch

Man

ager

Fina

nce

Man

ager

Fore

stry

Man

ager

Agr &

Adm

in

Man

ager

LST

Khar

man

gM

anag

erSh

igar

Man

ager

Skar

duM

anag

er

Man

ager

Mou

ntai

nIn

frast

ruct

ure

& E

ngin

eerin

g

Man

ager

Ente

rpris

eDe

velo

pmen

t

Man

ager

Lear

ning

Supp

ort

Unit

Man

ager

Soci

alDe

velo

pmen

t

Regi

onal

Pro

gram

Man

ager

FMUs

Mic

ro F

inan

ce

AK

RS

P

(M

ic

ro

-F

in

an

ce

)F

ig

ur

e

K.

3

Syst

em A

naly

st(C

ore)

PM M

icro

Fin

ance

Dire

ctor

Fin

ance

RPO

Chitr

alRP

O Gi

lgit

RPO

Balti

stan

Regi

onal

Man

ager

Mic

ro F

inan

ce

Mic

ro F

inan

ceM

onito

rLo

an P

ortfo

lioAu

dito

r (Co

re)

Mic

ro F

inan

ceM

onito

rM

F M

onito

r

Fina

ncia

l Ana

lyst

(Cor

e)

FMU

sFM

Us

FMU

s

Regi

onal

Man

ager

Mic

ro F

inan

ceCo

ordi

nato

r MF

1 0 1

Under the umbrella of the community organi-zations, the AKRSP provides support in the fourmain program areas discussed below: microfi-nance, productive physical infrastructure (PPI),natural resource management (NRM), and enter-prise development, together with the humanresource development associated with theseprograms. The evaluation team assessed each ofthe four program areas separately, again apply-ing the standard OED evaluation criteria. Thisannex provides the details of that assessment.

Microfinance

Relevance The objectives of the AKRSP Microfinance Pro-gram, as presented to the mission, are as follows:• Create a large capital base through regular

savings.• Provide sustainable access to microfinance

services. • Devise simple and appropriate financial man-

agement systems for village and women’sorganizations.

• Improve financial and managerial skills of vil-lage and women’s organization office bearers.

• Establish a sustainable financial institution.

Recent World Bank Country Assistance Strategies(CASs) continue to cite the importance of encour-aging NGOs and community-based organiza-tions to broaden service delivery and improveequality, and the need to increase agriculturalproductivity by improving the allocative effi-ciency of factor markets, including credit. TheAKRSP microfinance program is certainly con-sistent with this. However, the savings elementhas not always been an open savings program.Until recently, savings were compulsory, con-

tributed as the price of admission for commu-nity grants and individual loans backed by thesavings collateral. More recently, withdrawalhas been permitted.

Changing Circumstances Call for Clarificationof Strategy and Responsibilities. Total savingsdeclined for the first time in 2000, but rose mar-ginally in 2001, and the figure for loans disbursedhas been declining since 1997. The AKRSP isnow preparing for the establishment of a newnational-level microfinance bank, although thereare several steps to be completed before thisbecomes a reality. The AKRSP would be a majorshareholder in the new bank. The AKRSP’s pasthistory of changing credit products and its recentattempt at strengthening its financial managementsystem demonstrate responsiveness in the micro-finance program to the lessons of experience.However, changing circumstances and the pro-posal for the new bank are highlighting theneed to clarify objectives with respect to finan-cial sustainability and poverty alleviation strat-egy. Before moving forward, it will be importantto first redefine the basic objective, then rede-fine the target market and credit products andestablish the modalities of coordination betweenthe AKRSP and the new bank.

EfficacyMicrofinance Program’s Role in the Develop-ment of Village and Women’s Organizations.From the outset, the AKRSP has stressed savingsby community organizations as an investment insocial organization and as the capital base forinvestments in productive opportunities. Sincethe savings program has been designed to col-lect small amounts from rural households atregular intervals through community organiza-tions, this system has been very effective in

ANNEX L: MAIN PROGRAMS

reducing transaction costs and mobilizing sav-ings of poor households. While saving to a cer-tain level was compulsory, the microfinanceprogram was helped for some time by the dif-ferential interest rates and potential for arbi-trage—borrowing at low rates, saving at highrates. The program has demonstrated potentialfor capital accumulation and sound use of creditand has been instrumental in developing localcapacity in financial management. A total of Rs.429 million has been generated by the commu-nity organizations and a cumulative credit of Rs.1,459 million has been provided to 590,100 bor-rowers since initiation of the credit program in1983. More than 3,700 people have been trainedin bookkeeping and accounts, including manymanagers of community organizations.

Decline of Savings and Credits. Changes haverecently emerged that need to be addressed in thenew AKRSP strategy. Total savings declined forthe first time in 2000 to Rs. 429 million from Rs.432 million in 1999 (table L.1), although theyhave recently increased again. The leveling-off inthe rate of savings increase has largely comeabout because of the reduced volumes of com-pulsory savings as savings reserves have built up,and some competitive savings opportunities haveemerged elsewhere. On the credit side, whilethere had been rapid growth in the AKRSP’scredit disbursements from Rs. 1 million in 1983to Rs. 91 million in 1994, and to a high of Rs. 277

million in 1997, there has been a decline sincethen to Rs. 224 million in 1998 (loans outstand-ing: Rs. 253 million), Rs. 163 million in 1999(loans outstanding: Rs. 201 million), and down toRs. 127 million in 2000 (loans outstanding: Rs. 142million). This is partly because of the downturnin the economy and reduced demand and theAKRSP’s attempt to reduce portfolio risk, butmostly because of increased internal lendingwithin communities, although increased interestrates on all credit products have also contributed.

Increasing Internal Lending. Internal lending,earlier called village banking, is a self-managedcredit program. The community organizationlends its collective savings to the borrowers, whoare also its members, for productive and other pur-poses. Lending terms and conditions are set inter-nally by the community organization itself. Wheninternal loans are not repaid the communityorganization is faced with a difficult decision,either deducting the payment from the borrower’ssavings, which some community organizations arereluctant to do for fear of decapitalizing theorganization, or rescheduling the loan, whichmany allow, but which risks a decline in lendingdiscipline and erosion in the confidence of savers.Both of these outcomes risk disintegration of thesocial capital nurtured over many years.

Internal lending is an inevitable outcome ofincreasingly mature communities that feel moreconfident in self-management and have a greater

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

1 0 2

The AKRSP’s credit program, split into group and individualloans, meets the needs of a wide range of borrowers. It has pro-vided villagers with loans for agriculture and enterprise devel-opment as well as for consumption—particularly beneficial forwomen. The loan portfolios of the commercial banks are lim-ited. Most of them do not extend small loans. Commercialbanks do not have mobile banking services and most villagesare still outside the reach of formal banking. By 2001, there were28 branches of commercial banks and 21 branches of the Coop-erative Bank in the Gilgit and Ghizer districts. Commercialbanks have been able to accumulate an impressive amount indeposits. Three of the banks with the largest outreach in theNorthern Areas were found by the evaluation team to have

total deposits of Rs. 2.6 billion.a It was clear from interviewsin villages that there was an overwhelming preference forcredit from the AKRSP because of its timely disbursement andlow transaction costs. But the recent increases in interestrates have led to client dissatisfaction. There were also com-plaints of the single maturity date, and it was felt that the dura-tion of loans should be extended. The AKRSP has now madechanges in the single maturity dates in individual loans and hasallowed for payment in installments, so there is still a highdemand for AKRSP loans. However, internal lending has damp-ened credit demand.

a. National Bank, Habib Bank, and the Northern Areas Cooperative Bank.

T h e A K R S P ’ s C r e d i t P r o g r a mB o x L . 1

understanding of financial matters. It is esti-mated that in the Gilgit Region about 18 percentof the community organizations are undertakinginternal lending, and in Hunza, close to 50 per-cent. While this is in keeping with the AKRSPobjective of making villagers self-sufficient, thereare doubts about whether it will be sustainableat present village skill levels. Experience fromthe field shows large variations in the capacityof community organizations to manage theseactivities. The social dynamics of the organiza-tions are highly variable and it is not always pos-sible to secure the savings against default. Giventhe origin of the community organizations andtheir role in other activities, as well as the his-tory of compulsory saving by many poor peo-ple, the AKRSP may have lifted the controls onvillage lending too soon. While the AKRSP haslimited leverage in the internal decisionmakingof the organizations, it has been very active indeveloping policies and procedures to guidethe community organizations and the process.It has been promoting a set of uniform InternalLending Guidelines and providing training.1 This

effort is commendable, extremely important,and should continue. One option that has beenexplored is to have two community funds—one operating at higher risk and open to inter-nal lending, and the other one not.

Increased Level of Overdues and Write-Offs.Recovery performance has recently been underthreat, but the program should be able to recoverfrom this with the improved appraisal processesnow being applied. Recent figures show a highratio of overdues2 to loans outstanding—above20 percent, with a high of 25 percent in 1999—and a loan loss ratio increase to 4.1 percent in2000 from 1.7 percent in each of the two pre-vious years.3 Portfolio at risk4 peaked at 7.2 per-cent in 1999 (see table L.2). A review of loansoverdue reveals a marked regional difference inthe pattern of repayments. In 2000, BaltistanDistrict had the highest ratio at 48 percent, andChitral the lowest at 16 percent. A main reasonwas the substantial volume of risky lending forpotato production in Baltistan, where plant dis-ease and marketing problems affected repaymentcapacity.

A n n e x e s

1 0 3

Organization 1995 1996 1997 1998 1999 2000

Village organizations 199 248 281 318 327 319

Women’s organizations 48 62 75 96 104 111

Total 247 311 357 414 432 429Source: AKRSP Microfinance Section, data as of Dec. 31, 2000.

S a v i n g s B a l a n c e — C o m m u n i t y O r g a n i z a t i o n s( R s . m i l l i o n )

T a b l e L . 1

Measures 1998 1999 2000

A. Loans outstanding at the end of the period (Rs. million) 252.7 201.4 141.6

B. Loans overdue at the end of the period (Rs. million) 50.9 50.4 30.8

C. Portfolio at risk at the end of the period (Rs. million) 10.7 14.5 7.9

D. Loan loss reserve (Rs. million) 8.1 11.0 6.3

E. Overdue as % of outstanding (B/A) 20.1 25.0 21.8

F. Portfolio at risk as % of outstanding (C/A) 4.2 7.2 5.6

G. Loan loss reserve as % of portfolio at risk (D/C) 76.0 76.2 80.0Source: AKRSP 2000j.

P o r t f o l i o Q u a l i t y o f M i c r o f i n a n c e P r o g r a mT a b l e L . 2

Strengthening of Microfinance Management.Since the last OED evaluation, a number ofcommendable changes have been made in theoperations of the Microfinance Program thatshould improve overall efficiency and sustain-ability and help address the overdues problem.These changes include:• Upward revision of interest rates for all credit

products• Rationalization of credit products• Strengthening of the institutional capacity of

the Microfinance Section• Strengthening of the financial management

capability by introducing: 1) new monitoringreporting tools (AKRSP 2000j), 2) loan write-off and loan loss provision policies, 3) apolicies and procedures manual for the micro-finance operation, and 4) an automated man-agement information system (MIS).

• And, as the most fundamental change, seg-regation of the Microfinance Program, asautonomous operations from other main-stream AKRSP activities, from January 2001 inpreparation for the new bank.

EfficiencyOverall, efficiency is substantial when comparedwith Pakistan and global standards. While theoperating cost5 ratio over average portfolio out-standing increased to 11.3 percent in 2000 from7.1 percent in 1998, this is well below the indus-try standard of 13 percent to 21 percent (see table

L.3). The increase is a result of staff growth andthe improvement of the financial managementsystem. The AKRSP’s microfinance operations arenot directly comparable with most microfinanceNGOs in Pakistan because of the small size oftheir loan portfolios compared with the AKRSP,different implementation methodologies, andtheir limited time in operation.6 However, acomparison of the AKRSP with other NGOs inthe Microfinance Group indicates relative effi-ciency (see table L.4). The AKRSP’s operating costratio is among the lowest. But this is due to itslarge volume of performing assets, which mayhave reached a saturation point, while the vol-ume of loans may be growing for other NGOsrelative to their costs. Table L.4 also indicates thatthe efficiency of the AKRSP’s Microfinance Pro-gram is better than the industry average of 114microfinance institutions around the world.

Institutional Development ImpactThe institutional development impact of theAKRSP Microfinance Program has been sub-stantial and has meshed well with the commu-nity and investment activities. However, theproposal for the new bank presents a fresh chal-lenge, coming at a time when AKRSP microfi-nance is somewhat fragile.

Enhancing Local Financial ManagementCapacity. The AKRSP has played an importantrole in the training of village-based accountantsand managers. Nevertheless, there is further need

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

1 0 4

Indicator 1998 1999 2000

A. Operating cost (direct costs only) (Rs. million) 18.7 19.2 19.4

B. Average performing assets (Rs. million) 326 476 565

C. Average portfolio outstanding (Rs. million) 263 227 172

D. Operating cost ratio over average

performing assets: (A/B) (standard = 5 ~ 16%) 5.8 4.0 3.4

E. Operating cost ratio over

average portfolio outstanding: (A/C) (standard = 13 ~ 21%) 7.1 8.5 11.3

F. Cost per unit of money lent (Rs) 0.08 0.12 0.15

G. Cost per borrower (op.cost/no.of borrowers) (Rs) 551 719 894Source: AKRSP 2000j.

E f f i c i e n c y o f t h e A K R S P M i c r o f i n a n c eP r o g r a m

T a b l e L . 3

for strengthening their capacity to manage thefinancial systems at the village level, particularlygiven the emergence of internal lending. Moretraining of women managers is also needed toincrease their ownership and autonomy in recordkeeping and the management of village accounts.

Strengthening of the Microfinance SectionCapacity. The AKRSP has strengthened its Micro-finance Section—increasing microfinance officersat the field level from 6 in 1995 to 31 by the endof 2000. In addition, a loan portfolio auditor, afinancial analyst, and a systems analyst wereadded to its core staff. The AKRSP also createda separate cost center for the Microfinance Pro-gram and all direct and indirect costs werecharged to the Microfinance Program as a steptoward financial sustainability. As a result, theoperating costs of the Microfinance Programincreased from Rs. 2.8 million in 1995 to Rs. 19.4million in 2000.

The changes over the past three years aimedat portfolio quality improvement have led to anincrease in cost per unit of money lent, and anincrease in the cost per borrower (see tableL.3), while the average amount lent per borrowerhas gone down and the number of loans and thecredit amount being overseen by each credit offi-cer have decreased appreciably. This is likely toresult in some permanent increase in the cost oflending, but this is expected to be partly com-pensated by improved quality of lending in thefuture. The AKRSP is also now building a com-puterized management information system. It is

currently being tested in each Field ManagementUnit. Once fully installed, the system is expectedto help reduce the workload of the microfi-nance officers and improve the financial man-agement of the credit program. New reportingformats have been established to provide in-formation on efficiency, portfolio quality, andsustainability.

There are significant risks for poverty outcomein the shift toward Microfinance Section auton-omy, and finally the microfinance bank. Themove risks divorcing microfinance activitiessomewhat from the complementarity that otherprogram components could exercise on thecredit and savings portfolio.

In the past, there was thought that there wasan advantage in integrating the credit activitieswith technical assistance from other programcomponents. While under the particular cir-cumstances of the AKRSP’s unusually well-coor-dinated program this may have been the case,global experience has shown that this may actu-ally hurt credit more than help it, providing theopportunity for borrowers to default on thegrounds of faulty technical advice.

With separate institutions—the AKRSP and theproposed new bank—the advantage of securingloan repayments as an incentive for furtherAKRSP investments at the village level could beweakened, but the intention appears to be towithhold investments from defaulting villages.For financial sustainability—with potential pos-itive impacts for poverty in the longer term—the

A n n e x e s

1 0 5

MBBAKRSP peer group in MBB, Microfinance Group

Microfinance South Asia all participants in PakistanProgram (n = 9) (n = 114) (n = 10)

Operating costs

Average loan portfolio 12% 20% 31% 15%

Operating costs

Borrowers $16 $19 $150 $16

Borrowers

Staff 389 229 111 N/ASource: AKRSP 2000h.

E f f i c i e n c y C o m p a r i s o n sT a b l e L . 4

shift to a new bank is appropriate at this stageof the program. However, design of the respon-sibilities of the parties and relationships betweenthem needs to be carefully considered, and thefocus on the poverty objective sustained.

SustainabilitySustainability of the credit program over theyears has been good, with a 17-year record ofgenerally strong savings and loan performanceand modest levels of subsidy by industry stan-dards. With respect to financial sustainability, theAKRSP’s Subsidy Dependence Index (SDI) isestimated to be a very favorable negative 130 per-cent (Annex G), indicating no subsidy. However,three coefficients used in the calculation bias ittoward financial sustainability, depending onmethodology used.

First, the interest used is the concessionalPoverty Alleviation Fund rate. The AKRSP arguesthat it cannot borrow from commercial banks,so the Poverty Alleviation Fund (PAF) is itsappropriate opportunity cost. Nevertheless, it cur-rently has sufficient funds of its own and hasdone little recent borrowing, but in Pakistanthe opportunity cost of capital is estimated bythe World Bank at closer to 12 percent.

Second, other subsidies, in the form of oper-ational support from the rest of the program, aretreated as zero. The AKRSP argues that these arevery difficult to calculate since many are sharedwith the rest of the program, and that this shar-ing cuts both ways—non-microfinance AKRSPstaff and facilities help microfinance, but thereverse is also true. This is probably correct.

Third, the investment income from unusedfunds is included in the calculation. The AKRSPargues that, in accordance with the decision ofthe Board, these funds remain with the Micro-finance Section. This would not normally beincluded in an SDI. However, excluding theinvestment income and the equity involved alsopresents methodological problems. The right-hand column in Annex G adjusts for the PAFfunding only. This shows a virtually zero sub-sidy. Alternative, less favorable methodologicaltreatments still do not show high subsidy levels.

The income from service charges (interest)increased almost fivefold between 1995 and

2000. This is explained by the increase in inter-est rates, improved recovery performance, andthe increase in loan volumes. Another source ofsignificant increase in funds available has comefrom investment income. No income was beinggenerated from investments in 1995, but by theend of 2000, the AKRSP had improved its fundmanagement by investing its idle funds in high-yielding returns.

Recent Increase in Overdues. The increasedlevel of overdues, mentioned earlier, is largelya legacy of a particular recent credit operationfailure, mainly with individual loans (previouslycalled enterprise loans) for potato production inBaltistan, which faced production and market-ing problems. In the mid 1990s, the AKRSPattempted to rapidly expand its credit portfoliothrough experimentation with lending products,but without proper appraisal procedures.

However, the AKRSP has been proactive inreevaluating its loan products and quick to ter-minate poorly performing products. It has nowamended its internal policies and procedures. Inaddition, the AKRSP has introduced loan write-off and loan loss provisioning policies, and, atthe same time, intensified collection efforts at thefield level.

In January 2001, the AKRSP Microfinance Pro-gram started testing more systematized guidelinesfor delinquency management. This requires morefrequent contact between borrowers and micro-finance officers and is likely to raise costs, butthese costs are expected to be at least partly off-set by improved efficiencies from the new man-agement systems. Considering the regionaldifferences in the overdue pattern, an adequatelydifferentiated approach with particular attentionto the Baltistan District is required and appearsto be well in hand. Its effect on repayment per-formance needs to be closely monitored.

Sustainability and Establishment of a NewBank. The mission did not appraise the sus-tainability of the proposed new bank since NACactivities will only be part of the total business.We simply note that sustainability in the programregion of AKRSP—the NAC—is different fromsustainability of the new microfinance bank.Projections suggest that some cross-subsidy forthe NAC will be needed for the first five years.

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

1 0 6

Although the current AKRSP microfinance pro-gram has recently attained financial sustainabil-ity (or almost attained it, depending on how costsare calculated), the new bank is expected toincur a number of start-up costs.

Productive Physical Infrastructure The benefits and impact of the 2,000 small-scaleinfrastructure schemes completed so far are sub-stantial, far-reaching, and sustainable (see AnnexJ, tables J.6 to J.9). Half the schemes are for irri-gation development, almost a quarter are trans-port schemes (link roads, bridges, and tracks),and the remainder are a mixture of protectiveworks, microhydels, boundary walls, domesticwater supply, and other types of schemes, includ-ing plant nurseries, barbed-wire fences, sanita-tion and sewerage, and community bathroomsand centers. The schemes have already helpedtransform the villages that adopted them earlyand will soon do so for the villages that becameinvolved late. This is evident from a range ofAKRSP and independent studies, as well as fromthe evaluation team’s village visits.

Ingenious irrigation supply channels, oftenover long, precarious routes across mountainslopes, have developed 48,000 hectares of newland and intensified cropping on existing land.As is evident from a number of case studies, thishas resulted in impressive production increases.The hundreds of miles of link roads and 60road bridges constructed by the villagers since1983 have increased the impact of the main, gov-ernment-constructed trunk roads. The same willbe true for the AKRSP link roads and bridges inChitral and Astore as those main roads are sealedand improved. Microhydels are making a majordifference to life-styles in the region, especiallyin Chitral. The great variety of other types ofschemes chosen by villages is a sign of theopportunities for more diversified infrastructureinvestments in the future.

RelevanceThe objectives of the PPI program were to (i) pro-vide the essential catalyst for village organizationfollowing a period of organizational vacuum and(ii) respond with technical assistance and grantfunding to the expressed needs of the villagers

to improve their lives by investment in physicalworks. These objectives were highly relevant.Together with credit, infrastructure was a keysocial building block. The relevance and high pri-ority of the schemes was assured by the primacyof the village organization in the selection processand has been confirmed by high constructionstandards and generally good maintenance. Theobjectives are consistent with the element of theoverall mission statement directed at the provi-sion of economic resources and opportunities,although, again, equity is not addressed directly.

The infrastructure program has been highlyrelevant to the needs of the people of the NACregions. It has resulted in equitably distributedinvestments at what are claimed by AKRSP staffto be substantially lower cost than traditionalpublic works. The benefits are evidenced bystrong continuing demand in the villages formore schemes. Government has also joined theprocess by providing funds for extra schemes tobe overseen by the AKRSP. The schemes cho-sen by villages were a good match with tradi-tional construction skills acquired over hundredsof years in the harsh mountain conditions.7 Butrelevant to the question of incentives for sus-taining community organizations is that, in 1987,after 5 years of the program, about 20 percentof organizations had a PPI being initiated thatyear, whereas in 2000, with more than three timesthe number of organizations and some declinein number of PPIs initiated, only about 5 per-cent had a PPI being initiated that year.

EfficacySince the previous OED evaluation, the PPI pro-gram has continued to complete the long-termaverage of about 120 schemes a year, close tothe plateau reached in the late 1980s. After alower completion rate of 83 in 1999, programengineers were increased by 20 contract staff,and 154 schemes were completed in 2000. Thelevel of success over 14 years in sponsoring theconstruction of a broad range of small-scale,high-priority infrastructure projects has beenexceptional. The speed with which some proj-ects are completed is remarkable, with villagework crews contributing long days and camp-ing on-site when distant from the village.

A n n e x e s

1 0 7

Relatively few projects have failed outright.“Abandoned projects” totaled 84 at the end of2000, which is about 4 percent of completedschemes. (This was incorporated in the eco-nomic analysis.) Over 60 percent of these are irri-gation schemes in Chitral and Baltistan. Specialefforts are needed in those regions to reduce fail-ure of irrigation schemes, and possibly to resus-citate a number of them, given their sunk costs.Even so, 4 percent is commendably low forabandoned schemes considering the challengesin perhaps the most unstable rural topographyin the world and the exceptionally high level ofsocial cohesion needed—often between peopleof different sects—to mobilize the labor required.Problems leading to abandonment include socialdisagreements, especially over land (rights ofway, compensation, and allocations), and unfore-seen technical difficulties.

Efficacy of Irrigation Schemes. There is causefor concern about the rate of completion ofland development (tertiary channeling, rockclearance, terrace construction, and planting)to make early use of available irrigation water.Unlike the construction of the water supply sys-tem, this aspect is left to farmers and in someschemes has lagged, with a consequent reduc-tion in benefits. Reclassifying irrigation schemesas land development schemes represents a sig-nal, but it is not clear that this has translated intofaster completion. The evaluation team saw sev-eral schemes where it will be some years beforeall the available water is used.

Efficacy of Link Roads and Bridges. The com-pletion of the Karakoram and Skardu highwaysmore than 20 years ago much improved trans-port links. Similarly, the main highway improve-ments under way in Chitral and Astore willimprove transport links there. Astore in partic-ular seems to have exhibited low productivityand income gains due to poor infrastructure. Butthe benefits of these highways could not berealized without linking roads to the distantcommunities. Apart from some government con-struction of secondary roads, the AKRSP roadsand bridges have provided the bulk of the ter-tiary road system. They will continue to do soif identified demands for more roads can befinanced. The economic and social impacts have

been substantial, based on the economic analy-sis for this component, which suggests ERRsaveraging about 27 percent, and on village dis-cussions, which often stressed the health bene-fits. The evaluation team saw cases where alink road or bridge could reduce travel time todistant land or the nearest commercial centerfrom 2 to 4 hours on foot, to 20 to 30 minutesby truck.

Selection of these projects by the whole com-munity in open meetings conducted with AKRSPfacilitators generally ensures that the highest-priority schemes are chosen and that there willbe collective commitment to provide a sub-stantial share of the costs and to construct theschemes with their own labor. The evaluationteam did not come across a single road or bridgescheme where selection had subsequently beenquestioned. This is in contrast to many earlierpublic works that did not flourish because of site,design, construction, and maintenance prob-lems that could be traced to failure to consultlocal opinion on site selection and other tech-nical matters. Completion times have generallybeen excellent.

Efficacy of Microhydels. The benefits of micro-hydels are numerous and far-reaching. Waterpower is one of the few locally exploitableresources. Initially the main use is for lightingto replace kerosene lamps and wood fires. Thishas allowed children to spend more time study-ing and has permitted women to produce moretraditional crafts for sale. Smoke pollution fromburning wood and kerosene is reduced; basichousehold chores can be mechanized (such aschurning butter and washing clothes), releasingwomen’s time; communication is improvedthrough radios and televisions; and socializingis extended. In addition, the power has resultedin small enterprise development, such as stonepolishing.

Starting with small schemes in the early1990s that employed somewhat makeshift plans,the engineering skills of the microhydel teamwere rapidly strengthened, resulting inimproved scheme design and construction. TheChitral Regional Program Office (RPO) hashelped to install the bulk of the microhydels(153 of a total of 167 through 2000), complet-

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

1 0 8

ing an average of 17 schemes a year. Theschemes provide power to more than a thirdof the district’s 43,000 households. Unlike mosthydropower, these schemes do not requiredams for water storage, since they operate onwater from snow and glacial ice—natural formsof storage.

Exchanges of data with other microhydelprograms indicate that the Chitral region hasalready achieved the highest microhydel con-centration in the world, and more stations arebeing installed (see box L.2). There is scope inChitral for many more such schemes, for upgrad-ing existing ones, and for larger schemes toprovide a more advanced level of service formore power-intensive domestic uses, publicfacilities, agricultural processing, and small busi-nesses (DFID 1998). A start has been made inthis direction with the completion of a commu-nity washing station for 300 families and thedevelopment of low-wattage water heaters foroff-peak use.

The Chitral microhydel program has beenparticularly effective as a tool of social organi-zation, as well as providing the entry point forparticipation by communities that had previ-ously held back on sectarian grounds. AKRSPstaff have also provided assistance on the IFAD-funded Chitral Agricultural Development Projectmicrohydel component and a few private

schemes, as well as internationally, such that theAKRSP approach has reached well beyond theprogram.8

However, the success in Chitral highlights apublic policy issue that government needs toaddress. The rapid growth of schemes in the dis-trict is attributed to the high cost and unreliabilityof public electricity supplies in Chitral, com-pared with subsidized supplies in Gilgit andBaltistan. Cutting the subsidies in those regionswould open up more potential for microhydelschemes at no public cost.

EfficiencyIn early 2000, the AKRSP surveyed a small sam-ple of completed infrastructure schemes in the3 regions, 11 each in Gilgit and Chitral and 4 inBaltistan, which provided evidence to supportthe evaluation team’s positive observations(AKRSP 2000k). The main findings of the impact9

studies are summarized in box L.3 for the threemain types of schemes: land development, linkroads/bridges, and microhydels.10

With respect to allocative efficiency, evidenceis mixed. Average ERRs for the three main infra-structure types fall in a relatively narrow range,suggesting reasonably sound overall allocationby PPI type. However, the range of ERRs withinthe sample of cases is wider, suggesting thatwithin investment categories there may be room

A n n e x e s

1 0 9

Chitral has numerous sites for harnessing water power. Since1991, more than 170 schemes have been initiated by the AKRSPand about 150 have been completed, benefiting about 16,000households (37 percent of the 43,000 households in the district).The average AKRSP grant has been about Rp. 69,000 (US$10,600)per scheme, about US$150 per household. Water-driven grainmilling has been a tradition for centuries, but the AKRSP addedorganizational, engineering, planning, training, and implemen-tation skills to supplement traditional skills and provided cap-ital grants for materials and supplies. Water supplies are oftenplentiful, and schemes can be designed to be compatible withirrigation, since power generation only reduces the head, notwater volume (in the rugged terrain with many perennial torrents,sufficient head for power can usually be found without reduc-

ing the irrigation command area). The technology introduced hasbeen simple and robust, rather than sophisticated or highly effi-cient, suiting local capabilities. Training is provided for opera-tors, site supervisors, and for villagers on safety aspects.Advanced engineering training is given to selected supervi-sors. The strategy has included linking up with equipment man-ufacturers and nurturing the growth of a microhydel serviceindustry that will give additional benefits beyond the program.The schemes visited were more technically advanced than thefirst schemes seen in the mid-1990s. This is attributable to tech-nical assistance from Intermediate Technology Consultants,U.K., and links established with other agencies such as theInternational Center for Mountain Development (ICIMOD), Nepal,and with Powerflow, New Zealand.

M i c r o h y d e l S c h e m e s i n C h i t r a l D i s t r i c tB o x L . 2

for improved selectivity. The strategy is pro-poor, however, and allocating purely on com-parative ERRs may distort the pursuit of thatobjective. With respect to allocation within thevillage, the local contribution of about 30 per-cent is likely to elicit fairly careful considerationof perceived relative community returns.

An impact study of a women’s organizationwater supply scheme details outcomes similar tothose reported for other such schemes in the vil-lages visited (AKRSP 2000l). The main benefitsreported by the members were time saved car-rying water (three hours each day per family dur-ing the four months of winter); the time andmoney saved through reduced illness (less med-icine and emergency transport needed—abouthalf the usual expenditure); women had moretime available for socializing and for productivecraft work, and the improved health of childrenhad increased time spent on school work. Theestimated rate of return was satisfactory.

The one major reservation on irrigationscheme efficiency, mentioned earlier, is the longdelays in some schemes in completing the irri-gated area. Bundling land development in amore integrated fashion with channel or pipe con-

struction as a single scheme, on a planned sched-ule with credit funding, may improve efficiency.

With respect to comparative costs, an AKRSPstudy compared 230 projects constructed in1998/99 by region.11 The main findings were asfollows:• Costs per household were significantly lower

in Gilgit region at Rs. 4,800 (a ceiling is placedon family costs, and schemes tend to havelarger numbers participating), compared withmore than Rs. 7,000 for the other regions.

• Chitral has significantly higher unit costs peracre and foot of channel because of the lim-ited land available. On average, schemes pro-vided only about an acre of additional landper family in Chitral compared with 2.6 to 5acres elsewhere.

• Community contributions to costs were fromjust over 20 percent in Baltistan and Chitral,to as much as 43 percent in Gilgit region,reflecting a higher ability to pay in the moreprosperous Gilgit.

• Community contributions were highest forlink roads—the predominant public goodcomponent of infrastructure interventions—whereas equity and efficiency would sug-

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

1 1 0

A series of case studies by the AKRSP found that implementingthe projects increased social cohesion and that they had verygood rates of return, in the range of 25–35 percent.

Land development (irrigation) studies showed substantialcommunity contributions (16–46 percent); huge appreciation inland values (the evaluation team found even higher land valueincrements than the case studies); increased fodder produc-tion (restorative cropping of legumes) leading to increased con-sumption of livestock products; cereal cropping limited byinitially poor soils and subsidy disincentives; significant impacton fuelwood and timber resources; new land generally distrib-uted equitably; an increase of two hours a day in women’s work-load in the agricultural season; and good sustainability prospectsgiven villagers’ self-interest, maintenance committees, and col-lection of maintenance funds.

Link roads and bridges studies showed improved marketaccess (increased cash-cropping); time savings, land apprecia-tion; improved access by agricultural machinery (reducing workloads); some tourism increase; increased services from govern-ment and NGOs; easier access to the village for construction;

improved health service; maintenance is usually good, with someexceptions; where religious practices allow, there is increasedmobility of women; environmental impacts are mixed—con-struction losses of land and trees, stimulation of forest exploita-tion (an issue warranting increased attention in future planning),but reduced fuelwood use as alternative fuels become more read-ily available and cheaper; and, in some cases, increased use ofagricultural chemicals linked to their easier availability.

Microhydel studies showed that power is used mainly forlighting in place of kerosene (now a demand for upgraded micro-hydels to provide power for heating and cooking); benefitsincluded reduced kerosene fumes, more time for crafts, study,and entertaining; sustainability is good as schemes are wellmaintained (maintenance funds, operator training, and localcomponents), but there are sustainability concerns in somecases where flat rate charges are less than operation and main-tenance (O&M) costs where there are few meters and when thereis an absence of maintenance funds; reduced women’s work loadwhere household appliances are used; and equitable connec-tion and charges to all households.

F i n d i n g s o f I n f r a s t r u c t u r e C a s e S t u d i e sB o x L . 3

gest that public goods should enjoy the high-est subsidy level.

Institutional Development ImpactThe infrastructure program has had substantialinstitutional development impact as the main cat-alyst for social organization. It has enhancedsocial cohesion in villages (and between vil-lages where cluster PPIs have been undertaken)and probably contributed to equity mechanismsthrough the associated land allocation proce-dures. The implementation and management ofinfrastructure has led to strengthening of skillsin community organizations and led to wide-spread recognition, including by government, ofthe effectiveness of community organizationsas the village operational arm.

SustainabilitySustainability for the infrastructure schemes isassessed as highly likely, but with reservationsfor some types of schemes. Village ownershipof schemes is excellent. Schemes are identifiedas high priority, are constructed by the village,and provide substantial benefits to most, usuallyall, members. Dissension over operation andmaintenance (O&M) of facilities is relativelyrare. In many cases the village organization hasspun off scheme committees to handle suchmatters as maintenance, which has strength-ened sustainability by placing specialists incharge of schemes.

Technical soundness is exceptional as thedesigns and construction methods combine con-siderable traditional knowledge with the exper-tise of experienced AKRSP staff. Where technicalproblems have arisen, ways to resolve themhave usually been found quickly.

Maintenance of infrastructure is of a gener-ally high standard: the AKRSP reports that 92 per-cent of the infrastructure projects are effectivelymaintained, with a high figure of 98 percent forthose in Gilgit region (AKRSP 2000a, p. 19, sec-tion 4). Since the 1995 evaluation, the intro-duction of mandatory maintenance funds andcommittees has strengthened maintenance.Although the evaluation team’s resources did notallow extensive checking of maintenance, noserious problems were reported during village

meetings, and no major deficiencies wereobserved at selected projects.

This is an exceptionally high level of per-formance in the rural development field, espe-cially for scattered small-scale construction inremote areas. In many rural development pro-grams maintenance is not only weak but also isoften a threat to scheme and program survival.The achievement is more remarkable in NACbecause the predominant works (irrigation chan-nels/pipes and roads) are highly prone to seri-ous damage from mud and rock slides.Maintenance standards of private, or small group,assets can be expected to be quite good, but herethis is also true of public goods where there isa free-rider option, such as roads and bridges.The success can be attributed to the self-interestof villagers in continued benefits, identificationof the villagers with the structures, the AKRSP’ssupport for social organization, careful screen-ing of scheme proposals for widespread support,the relatively equitable benefits, the modestlevels of exotic technology over traditionalknowledge, the good training of operators; and,more recently, the insistence by the AKRSP thatmaintenance funds are set aside by communityorganizations.

However, there are three sustainability issuesof some concern:• Cost recovery in microhydel schemes is only

just sufficient to cover routine O&M, with noreserve for larger repairs or replacements.Currently, ad hoc levies are used for routineO&M expenses. While such a system maywork for small schemes, the pattern of low-cost energy this has set cannot be sustainedin the planned expansion to larger, more effi-cient schemes. The RPO in Chitral is awareof the problem and action is being taken toincrease tariffs gradually to more sustainablelevels (a similar problem is likely for domes-tic water supply).

• In some irrigation schemes there should begreater attention to projecting labor con-straints. In some areas, there are increasing,competing employment opportunities. Theplanning of irrigation where land expansionis high relative to resident adults (some newschemes have more than doubled land area)

A n n e x e s

1 1 1

should take careful note of seasonal labordemand and supply. Labor availability is amain reason why it takes a number of yearsto finish land development following thecompletion of an irrigation channel.

• Water use efficiency is becoming increasinglyimportant. In some villages there has beendecreasing water—possibly associated withclimate change through lower precipitationand consequent lower glacier and snow melt.While this is not yet widespread, and thereremain quite a few unexploited water sources,if the declining trend continues and compet-ing water uses continue to rise, the sustain-ability of some future irrigation investmentscould be threatened.

Natural Resource Management

RelevanceNatural resource management (NRM) activities,encompassing agriculture, livestock, and forestry,have been highly relevant and have had a sig-nificant beneficial impact in the project area overthe period of the program, although quantifica-tion and attribution are difficult. NRM activities canbe expected to remain relevant as the bulk of theincome of most households (about 60 percent ofhousehold income comes from farming).

The NRM strategy calls for raising the standardsof living and incomes of the people of the North-ern Areas and Chitral to a level comparable withthe national average for Pakistan. It proposesfour priority areas: increasing area under irriga-tion; wheat12 and maize improvement; animalhusbandry; and commercial agriculture. The strat-egy also proposes three strategic elements, as anoverlay: increasing production from existing farm-land by 66 percent; increasing cultivated area by30 percent; and modestly increasing incomethrough ancillary enterprises such as beekeeping,poultry, and trophy hunting. Increases in agri-cultural production are expected to come fromgreater cropping intensity, new technologies, andimproved supply of inputs. Forestry is beingplayed down somewhat, and will focus on newland only, because of substantial planting achieve-ments and an increase in private nurseries initi-ated with AKRSP assistance.

One premise of the strategy—that productionmust increase by 66 percent in line with popu-lation growth (2.55 percent)—is inappropriate.It implies that agriculture must maintain its cur-rent relative contribution, which is not self-evident for an area with such special resourceand logistical constraints and with the availablealternatives for raising household income. Thusimproving livelihoods and raising householdincome would be a more suitable target andwould direct attention to non-farm income. Withrespect to the first of the four thrust areas, thesemi-formal questionnaire applied by the teamat the village level, together with field observa-tions, confirmed the potential for expansion ofirrigated land area, but to fully realize thesebenefits speedier land development is needed.

Efficacy Efficacy of the NRM component over the fullperiod of the program has been highly satisfac-tory; more recently it has been only satisfactory.Over the only period for which good income dataare available, per capita farm incomes increased2.7 times, from Rs. 2,647 in 1991 to Rs. 7,046 in1997 (in real terms, 1999 prices). The lowestpercentage increases were in Astore, where infra-structure is a main constraint, and the highestwere in Chitral (AKRSP 2000c). Efficacy was par-ticularly impressive in forestry. As noted by the1997 Joint Monitoring Mission, there has also beenan increasing environmental awareness in villagesover recent years, and some villages now con-trol free grazing—with significant potential farmsystems and crop management impacts—andattempt management of the high pastures. Therehas also been an increase in AKRSP attention toindigenous knowledge. However, the evalua-tion team had some doubts about whether thepoorest farm households are really receivingappropriate technology options explicitly devel-oped with very low input costs, quick returns,low risk, and labor requirements that eitheraccommodate off-farm work or provide signifi-cant, value-added employment. The need is forsomething similar to the targeted poverty pro-gram, with differentiation that would identifythe particular needs of the poorer households ineach identifiable zonal system.

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

1 1 2

Efficacy has been impressive in cash crops,but these gains will prove fragile if more atten-tion is not given to addressing crop and livestockpest and disease problems, understanding andlinking marketing and production, and enhanc-ing the capacity of the private sector input sup-ply system. Lack of appropriate technologiesfor a number of situations continues to be oneof the principal constraints on further growth.

Improved varieties of wheat and maize, 400metric tons of which have been distributed sinceinception of the program, is acknowledged byvillagers to have contributed to raising overallcereal production. Well over 50 percent of house-holds benefit from wheat technology (in Gilgitand Baltistan regions), and about 40 percent ofhouseholds benefit from new maize varieties.Apart from cereals, which are seen more asfood security crops than cash crops, growth infruits, vegetables, and livestock has significantlycontributed to increased farm income.13 Intro-duction of exotic varieties of cherries, apples,pears, and apricots; aggressive promotion ofpotato cultivation; and continued attention toincreased forage production (alfalfa, oats, maize)are examples of relevant and successful tech-nology innovations supported by the AKRSP.

However, there are some signs that increaseson both the input and output sides are reach-ing a plateau. In double-crop areas productiv-ity increases are leveling off. With so muchgenetic material introduced over the years, it wasdifficult for farmers to recall benchmarks againstwhich to measure change. Farmers in a numberof villages visited perceived a decline in soil fer-tility. The evaluation team observed some vari-ability in the performance of NRM interventionsacross villages and districts, which suggests theremay be room for management improvements.

Has There Been Some Loss of Vision in Recent Years?While it is difficult to prove, the evaluation teamsensed some loss of vision in recent years in nat-ural resource management, although perform-ance remained satisfactory. Evidence for thisincludes the weak explicit linkage between NRMstrategy and poverty objectives, the continuedlack of attention to the above-channel areas

and associated systems issues other than theongoing study, and, more broadly, the limitedresponse in the farm systems direction. As notedby the 1997 Joint Monitoring Mission, there hasbeen a lack of systematic, long-term approachesin NRM. But this finding should not be over-stated. The NRM program is still quite strong. Thisloss of vision may be better explained as a fail-ure to gain vision, arising from the shift from theseparate agriculture, livestock, and forestry activ-ities, with relatively simple but unlinked strate-gic concepts, to the largely donor-triggerednatural resource management concept—whichis more difficult to articulate and implement butoffers potentially higher rewards.

Strengths and WeaknessesThe evaluation team’s indicative village surveysprovided additional information on agriculturaltechnology and the AKRSP’s responses to prob-lems. About 60 percent of NAC farmers perceivethat cereal yields have stagnated in the past fewyears, and when asked if the AKRSP has doneanything to respond, 61 percent of them said“no.” With respect to livestock systems, only 8percent of the technologies perceived to bemost beneficial were related to livestock, andabout one-third of groups interviewed said thatlittle had been offered on livestock technologies,although livestock represents about 40 percentof farm income. However, the low percentagehere may be a product of farmers not perceiv-ing “new technologies” in livestock in quite thesame way as an improved crop variety. In aboutthree-quarters of villages visited, the evaluationteam rated overall AKRSP responsiveness to thelocal farming system as “responsive” or “veryresponsive,” although some of the “responsive”ratings were qualified as “only just.”

The evaluation team assessed strengths andweaknesses with respect to five attributes gen-erally considered important for technology devel-opment and dissemination. The findings aresummarized in table L.5.

On another dimension, with respect to sevenimportant ingredients for technology genera-tion, the evaluation team judged three areas ofthe AKRSP’s NRM capacity—staff skills, thecapacity to access available technologies, and the

A n n e x e s

1 1 3

evaluation of available technologies—to be rel-atively strong. It found four areas—mobilizationof farmer knowledge, the availability of processesfor identifying productivity constraints, farmerinvolvement in the identification of technologies,and dissemination and input supply—to be lessstrong.

The Farm Systems Recommendations of thePrevious Evaluation The evaluation mission’s view is that the farmsystems recommendations of the previous mis-sion should have been implemented (see boxL.4), although there may be room for debateabout the appropriate intensity and scale. Thereremains a case for implementing them, but withsome additional focus on a broader livelihoodsapproach that incorporates increased emphasison off-farm income sources. Focus on this lat-ter area should now lie within the remit of theNRM component since there is no other suitablehome for it and it is inherent in a systemsapproach. Moreover, it has generally been foundthat farm income is positively, not negatively,associated with non-farm income—that is, they

are complementary. The need is more for under-standing the processes and linkages14 and beingable to locate these in a recommendation domaindefined both zonally and by household resourceratios. This is more than simply zonal mapping.It differentiates by such factors as householdtype, landholding size, poverty level, labor avail-ability during peak season, level of livestock con-trol in the village, and marketing constraints. Itis still not entirely clear why the recommenda-tions were not adopted. It appears that the rea-son is a combination of concerns about costs;lack of a comprehensive NRM strategy; lack ofcore leadership in NRM; and lack of strong man-agement commitment to NRM relative to otherpriorities.

The system links to marketing, in particular,need to be explored.15 A large number of tech-nologies have been tested in the NRM program.A number have been successfully introduced andexhibit rising adoption rates, but there havebeen quite a few unsuccessful interventions,and some, especially in the livestock sector,where the farm system or marketing reasons forthe lack of adoption are not entirely clear. A

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

1 1 4

Attribute Strength Weakness

Integration All disciplines of crops, livestock, and forestry Limited central leadership in NRM to play mentor as

integrated into NRM; dissemination processes well as strategic planner’s role.

devolved to regional program offices.

Flexibility Regional management empowered to make Most NRM staff still target-oriented; little time left

periodic reviews and change, when needed, for creative activities or pursuing ideas that would

resources allocated for specific activities; can lead to generating new technologies.

be very responsive to community needs.

Relevance Several examples of relevant and successful A few areas/farming systems still do not have

technologies disseminated by the AKRSP can appropriate technical packages and associated

be found in the field. support (single-crop zones/above-channel farming systems;

remote communities with marketing problems).

Partnerships Very strong in building partnerships with Relatively weak in promoting partnership with government

communities. Emerging capacity in building agencies.

partnerships with private sector input suppliers.

Institutional support Over the full program, solid support from Support for NRM declined somewhat in recent years.

management; transparent policies;

goal-oriented.

S t r e n g t h s a n d W e a k n e s s e s i n N R MT a b l e L . 5

greater systems perspective would have helpedand has the potential to reduce technology ini-tiative dead ends. The true net costs may, there-fore, be modest.

Livestock Development in the Above-ChannelRange AreasWhile the AKRSP’s role has been importantthrough the afforestation program and throughfacilitating contacts between communities andNGOs such as the IUCN and the World WildlifeFund (WWF), in livestock productivity the AKRSPhas focused almost exclusively on the below-channel area. Yet nearly all the weight gain ofmost livestock comes from above-channel forage,and livestock is the biggest component of farmincome. This calls for more attention. In fact,many village livestock barely survive the winter.

The above-channel areas16 represent a com-plex system of natural forests, high pastures,summer livestock camps, and sparse wildlife, allscattered across an unstable landscape. The sta-bility of these fragile areas is closely linked to thebelow-channel farming systems through the graz-ing needs of village herds and flocks. Once theongoing research program covering these linkagespresents findings, the AKRSP should revisit itsdecision to stay away from the above-channelareas.17 If the decision is made to focus more onthese areas, then the AKRSP should collaboratewith the IUCN and the WWF. A shift in this direc-tion by the AKRSP would complement the effortsof these two agencies. It would broaden the sys-tems perspective in the direction of productivitywhile maintaining an environmental sustainabil-

ity and biodiversity focus, which, in any case, hasa substantial income potential through tourism andselective sport hunting. Indeed, in the longer-term,these high pastures are the leading areas of NACcomparative advantage. The AKRSP should bemore involved.

Efficacy of Government ResearchWhile not an AKRSP task, as discussed in Chap-ter 5, it will be important for the NAC that thegovernment increase its capacity in research.There is some evidence of recent improvementsat the Karakoram Agricultural Research Insti-tute for the Northern Areas (KARINA) (Annex I).If this can be sustained, perhaps with donor assis-tance, it may provide a basis for greater part-nership between government and the AKRSP.

Efficacy of Subcomponent InvestmentParticipatory Variety Selection (PVS) Program.While highly relevant, it is too early to assess theefficacy of this program, which is in its firstyear. The collaborative nature of PVS, involvingtarget germplasm from national (such as NARC,Pir Sabak) or international research institutes(such as the International Center for Maize andWheat Improvement—CIMMYT), can be usednot only to identify new cereal varieties, but alsoto test new agronomic practices, for example,seed priming or finding optimal fertilizerrequirements by soil type, and also for newcrops, such as triticale, rye, or oats. The programis concentrating first on wheat, and has yet tocommence for maize. It will require moreresources for scaling up.18

A n n e x e s

1 1 5

The 1996 evaluation recommended the following: (i) categorize thefarming systems so that relevant technology can be developed anda more informed dialogue carried out with communities; (ii) focuson clusters of villages and design the research on the basis of thoseclusters; (iii) have master farmer representatives (two per villageorganization) from both the wealthy and poorer categories ofhouseholds to accommodate differences in technology needsand risk resilience; (iv) use one site for each ecological zone, butalso ensure that each zone is covered in each of the three regions;(v) work at each site on technologies and their linkages, market

opportunities, land and water, and inter-household differences andwomen’s programs; (vi) link with local public and private researchinstitutions in each region (some technical assistance was pro-posed to help establish participatory farmer/researcher processes);(vii) further train AKRSP professionals in a systems perspective;(viii) village specialists and master trainers would continue to playtheir role as would the FMU; and (ix) collaborate with national cen-ters in testing a farm-based interface between small farmers andresearch aimed at influencing research programming at Jaglote,Chitral, and Skardu.

F a r m S y s t e m s R e c o m m e n d a t i o n s f r o m t h e1 9 9 6 O E D E v a l u a t i o n

B o x L . 4

It may take several years, but productionincreases in the range of 10 to 40 percent canbe anticipated with an effective seed multipli-cation and distribution program. Over the longerterm, a shift toward a participatory plant breed-ing (PPB) program within a government part-nership offers potential for wheat in thesingle-cropping zone. Because few, if any, suit-able varieties are likely to be available off theshelf, given this zone’s special characteristics.19

While the participatory aspects of such a programlie within the AKRSP’s role, the technical plantbreeding lies outside it. Plant breeding, whetherbasic or adaptive/participatory, should remainthe domain of the public sector, using Jaglote andsubstations, but with the AKRSP as a valuablepartner in the participatory approach.

Seed Supply. The AKRSP has been very effec-tive in disseminating seeds (Annex J, table J.1shows that 890 metric tons of improved seedshave been distributed, with Baltistan taking morethan half). This was unavoidable in the early yearsbecause there were no commercial suppliers.Commercial seed for self-pollinating crops suchas wheat, for which farmers recycle seed and forwhich there are low margins, is unlikely in thenear future. For these, the AKRSP is currently stillthe main vehicle and will need to continue to beso for several years. But priority should now begiven to supporting the emerging seed industrywith training and investment advice, as withcommercial vegetable seeds through North-SouthSeeds Company, an AKRSP-sponsored privateenterprise in Gilgit. In the future, the AKRSPshould confine itself mainly to training and sup-port for suppliers, and farmer training on selec-tion and storage. Direct seed supply activities maystill be justified on poverty grounds in the poorerhigh-altitude areas where government is lesslikely to have resources to operate. The targetwould be cold-weather vegetables, cold-tolerantfodder crops, kales, and root crops.

Seed subsidies continued too high for too longwithout a clear rationale. By 2000, 44 tons of veg-etable seed and 243 tons of potato seed, for anentirely commercial crop, were distributed at 50percent subsidy. This was substantially reducedafter 1998/99. The NRM group will need tofocus on finding ways to sustain seed produc-

ing groups with declining or no subsidies. Forcross-pollinated and non-grain crops such asmaize and forage (for which seed production ismore specialized and commercial) the AKRSP hasalready successfully used a farmer groupapproach.

Single-Crop Areas. The AKRSP has somewhatincreased its attention to single-crop areas at thehigher altitudes through breed improvementand fodder security activities. But these areasremain a challenge and a priority. They includethe poorest villages and most of Astore, whichhas had only modest net farm income gains. Aspart of a push on poverty-related NRM inter-ventions, these areas need a special program andstrategy with targets and monitoring. Expandingthe range of improved varieties, especially in fod-der, would be a central element.

Afforestation. The forestry program of theAKRSP has been a remarkable achievement.About 40 million trees have been planted withabout 70 percent survival, and about 1,500 pri-vate nurseries have been established. The eval-uation team saw many good plantings that confirmthis. The impressive gains are partly attributableto a collaborative effort with a Norwegian Agencyfor International Development (NORAD)–fundedsocial forestry project in the Northern Areas.Nonetheless, the program must prove its sus-tainability once subsidies are removed. Areas forpossible attention are commercialization of treeplantations through refined skills in nursery man-agement (the mission noted, for example, thatnursery root trainers, which have given impres-sive results elsewhere in dry areas, have not beenused); cost-effective transportation and trans-planting of seedlings; integration of social andtechnical aspects of plantation management; andtree felling/disposal and marketing of produce.It may be necessary to diversify away from whatis currently a predominantly poplar culturebecause of the risk of losses from borer attacks.There may also be opportunities for gains inunderstanding intercropped forage competition(notably in the traditional willow plantations), andthe economically optimal productivity cycles forpopular fodder species.

Livestock Husbandry Support. While supportin this area has been quite strong, improved feed

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

1 1 6

efficiency for ruminants is an area where theAKRSP’s interventions have been less success-ful. This is surprising given the importance oflivestock in the farming system. Adoption ratesfor two main items of technology, silage and ureastraw treatment, show poor uptake despite sub-sidies, although a third area, manger-based stall-feeding to reduce feed loss, is faring somewhatbetter. While labor constraint is generallybelieved responsible for the low uptake in thesetechnologies, there may be a case for a farmsystems–based analysis of the issue. It is prob-ably more complicated than just labor, possiblyinvolving pricing relationships between foragefor sale and feed, gender issues, and links tomarketing.

Veterinary Input Supplies. Veterinary suppliesshow an impressive record of distributionthrough the AKRSP’s network. However, con-tinued subsidy on such inputs well into the late1990s was a questionable strategy. Training andsupport for village livestock specialists and mas-ter trainers was partly directed toward veterinaryservices privatization, mainly for vaccinationand medicines supply. The role played by theEnterprise Development Section and the NRMstaff in facilitating establishment of veterinarysupply businesses has been a commendableexample of the sort of private sector partnershipthat should continue to grow in the future. Butthe AKRSP still accounts for substantial directsales (50 percent in two major stores in Gilgit,for instance). This should be phased out. TheAKRSP should continue its awareness-raisingrole with government, in particular, supportingregulation on the quality of inputs and trainingthe private sector and village-level workers.

Breed Improvement for Large Stock. This activ-ity has been effective after some difficulties withadaptability of breed choice. Crossbreeding ofcattle has resulted in significant increases inproductivity and head count. The crossesbetween Friesian bulls and local cows gave wayto preferences for the hardier Jersey bullsbecause the offspring produced more milk withhigher butter content. The adjustment is a goodexample of the AKRSP’s flexibility in keeping theprogram relevant and effective. Baltistan region,where the program is more successful than else-

where, is said to now have a substantial JerseyF1 population of about 7,000 head. Anotherway forward initiated in Gilgit is to select andmaintain elite herds of local animals with selectedfarmers, while some farmers maintain pure Jer-sey herds to produce stud bulls—currentlybrought from down-country. A challenge, how-ever, is stabilizing the crossbred progeny bymore systematic tracking of the successive prog-eny and monitoring of their performance.

Poultry Improvement. This program has notbeen as effective as the breed improvement ofruminants, partly due to technical problems andlack of a clear strategy. Problems were com-pounded by a lack of breed evaluation—com-mercial and farm-level rural production systemsrequire different birds. Evaluation usually can becompleted in two years by monitoring per-formance in selected villages, and chick pro-duction can then be promoted through trainedproducer groups in larger villages or throughcluster village organizations where there is amarket. The distribution of indigenous breeds toremote areas in Gilgit and Baltistan regions (Chi-tral may start soon) has merit for the povertyobjective and as a source of income for women.20

Efficiency Overall, efficiency appears satisfactory. The eco-nomic analysis of the NRM component alone,excluding the AKRSP overheads, which havebeen applied to the overall economic analysis,suggests a rate of return of about 25 percent. Thisrate of return is fairly typical of NRM interven-tions with high-technology content. NRM-relatedexpenditure per household declined from Rs. 762per household in 1998 (Rs. 80 million/105,000households) to Rs. 630 in 2000 (Rs. 70 mil-lion/111,000 households)—a 20 percent declinein real terms. The NRM program as a percent-age of the operating budget has fallen steadilyfrom 18.6 percent in 1998 to 15.4 percent 1999,13.6 percent in 2000, and 12.6 percent in 2001.Cutting down on input distribution subsidiesseems to be one cause of the decline and wouldalmost certainly represent an improvement inefficiency. Decreasing emphasis on input sub-sidy and distribution-related expenditure in favorof increasing emphasis on technology genera-

A n n e x e s

1 1 7

tion and dissemination, farmer training, and mar-ket information are likely to improve efficiency.

Institutional Development ImpactIs the AKRSP Substituting for Government inResearch and Extension? While, in the absenceof a control, it is difficult to prove that theAKRSP has substituted for government, the totalabsence of government extension found at thevillage level suggests that the AKRSP has prob-ably substituted for government in technologydissemination work. In research, with the AKRSPonly handling such adaptive work as testing/demonstrating new varieties, the AKRSP contri-bution has been more complementary.

Overall, had the AKRSP not been present, gov-ernment alone would certainly not have achievedanywhere near the results evident, althoughthere have been some government achieve-ments in technology. For example, some high-yield wheat varieties found their way into theNorthern Areas at the onset of the Green Rev-olution through the help of government, and aTibetan wheat variety—still popular in most ofthe single-crop areas—is said to have beenbrought by a Department of Agriculture officialwho visited Ladakh in the mid-1980s. But suchintroductions have been sporadic. The without-AKRSP scenario in NRM seems likely to havebeen, at best, one of low-level technology, prob-ably losing ground against national averages.

Master Trainers and Village Specialists. TheAKRSP’s present training programs are morethan adequate in quantitative terms—averagingabout 2,000 regular and 1,600 refresher coursesa year (Annex J, table J.10). In the survey of thevillage by the evaluation team, 91 percent ofrespondents said that village specialists are effec-tive, of which 24 percent said they are veryeffective. While such responses to an informalsurvey carried out with others present need tobe interpreted with caution, quality of skillsseemed to be appreciated at the village level.However, an M&E survey in Chitral, based onpreliminary findings, found that the attrition ratefor master trainers could well be over 40 percent.Attrition losses must be expected since mastertrainer service is entirely voluntary, but it is notall loss since such skills will still have an impact

through less formal channels. This may changein the future as greater specialization anddemand for skills meet an increasing readinessto pay for training. Meanwhile, categorization ofmaster trainers would enable greater trainingfocus. Increasingly customized training will beneeded to create the demand and a gradualshift toward payment for advisory services. Asparticipatory research becomes more estab-lished, the village specialists and master train-ers should become coordinators, operators, andcollaborators and will need training in this area.

The AKRSP should consider studying theincentives for training with the objective of rais-ing household financial returns to investing timein increased skills, while also gradually raisingcost recovery from both villagers and trainees.

The proposed new NRM strategy places agreater emphasis on technology. With increaseddemand for high-value commodities—includ-ing livestock products—that are technically morechallenging, training programs will need to befurther adjusted, probably toward an increaseddegree of specialization. The many matureAKRSP village organizations capable of seekinginformation should enable advisory services tobecome both more pluralistic and more com-mercially oriented.

Overall NRM support should be differentiatedbased on the assessed capacity of each villageto manage its development affairs and technol-ogy needs with less outside assistance. TheAKRSP’s recent efforts to promote input suppli-ers and phase out input subsidies supports thisapproach. Input suppliers should, with training,gradually assume more of the role of extensionagents, although some quality rating/stamp ofapproval for suppliers may be helpful initially.

With respect to the AKRSP staff, recent reviewssuggest that significant NRM training is requiredto orient field staff to the roles proposed for themin the new strategy. Two topics recommendedin the NRM strategy background documentswith which the mission concurs are: methodol-ogy for research-related activities and advisorytechniques aimed at a farm management busi-ness approach. The objective should be toenhance the skills of some staff from traditionaltechnical specialists to interactive development

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

1 1 8

facilitators who can help farmers graduallydevelop the skills to seek out and test their ownsolutions.

Sustainability

Financial Sustainability and Level of CostRecovery in NRMFinancial sustainability is improving. The AKRSPhas been increasing cost recovery on most inputsand has reached zero subsidy on many items (seetable L.6). Some elements of subsidy are con-sidered justified and have been retained, suchas a 30 percent subsidy on vegetable seeds inthe resource-poor areas. But the current direc-tion in the program is strongly toward increasedcost recovery. No fees are yet charged for AKRSPmaster trainer or village specialist advisory serv-ices, except for service fees for vaccinations byvillage specialists.

Sustainability and Self-SufficiencyIn an area as remote and challenging as the NAC,sustainability does not call for regional foodproduction self-sufficiency. Income transfer andnon-farm income represent an efficient element

in a rural areas sustainable livelihoods strategy.Food import may always be needed, paid forthrough earnings from elsewhere.

Environmental Sustainability There has been no broad environmental studyof the environmental costs and benefits con-tributed by the people of the NAC to this veryimportant river basin. Such a study is needed toassess the net services provided, but it wouldneed to be linked with national work. Withrespect to accelerated soil loss arising from landuse practices, sustainability issues do not appearto be a major concern in comparison with thevast movements of material arising from naturalweathering and collapses in this geologicallyyoung mountain region.

In the steep terrain typical of the region, withrising mountains and huge annual snowmelt,large quantities of both water and material arecarried down into the valleys. Indeed, the devel-opment of terraced land relies on the controlledharvest of that material. Similarly, the sealing oflong irrigation supply channels depends on siltdeposits early in the season. There are undoubt-edly some locations where, at the margin,

A n n e x e s

1 1 9

Current (2000–01)Intervention Before 1998–99 Average areas Resource poor areas

Seeds, wheat and maize 50 100 100

Seeds, fodder 50 100 100a

Fodder, block plantation 50 50b 50b

Potato seed 50 100 100

Vegetable seeds 50 100 70

Fruit tree seedlings 50 87b 87a,c

Forest tree seedlings 0–50 67 50

Livestock, bulls and rams 50 50 50

Livestock, cows 50 100 100

Livestock, poultry 70d 100 100a

Fisheries, fingerlings 50 100 100a. About 10 percent of households are supplied with 100 percent subsidized inputs.

b. Limited to a maximum of 10 hectares.

c. Only transport subsidy of about 13 percent.

d. Transport and mortality of chicks were subsidized.

L e v e l s o f C o s t R e c o v e r y ( p e r c e n t )T a b l e L . 6

improved land management would enhance pro-ductivity by either holding soil for longer wheremore is needed, or by helping to get it off fasterand in a managed fashion where silt is damag-ing.21 But these appear to be mostly private orlocal community land management decisions.While interventions such as afforestation cantemporarily detain more soil, and while indis-criminate tree cutting arising, for example, fromill-considered new roads, can do the reverseand cause significant biodiversity loss, the like-lihood is low of human activity raising or low-ering the total silt load coming out of such areasby appreciable amounts. Nevertheless, a betterunderstanding of the broad environmental rela-tionships is warranted.

With respect to road construction impact onforests and biodiversity, there have been somenegative impacts. If the AKRSP moves towardlarger multi-village infrastructure projects, envi-ronmental assessments will be needed.

With respect to pesticide use and fertilizerrunoff, there are no data. Training to prevent theindiscriminate use of inputs and further work onintegrated crop management techniques shouldbe pursued as much for profitability as for envi-ronmental reasons. However, the scale of fertil-izer and pesticide use is still quite modest andunlikely to present a major problem yet. But theissue will need careful watching. The 1997 JointMonitoring Mission proposed the developmentof environmental Safe Minimum Standards (SMS)in areas such as land use, soil conservation,grazing management, and sustainable tree har-vesting. While increased understanding at the vil-lage level of cropping and grazing systeminteractions with the physical environment isimportant and could contribute to increasedincomes, we have doubts about the value of anSMS approach. We doubt that the technical rela-tionships and their interaction with social factorscould be adequately understood. In drylandpastoral grazing systems in Africa, minimumstandards have proved impractical and oftenmisleading. However, a holistic resource man-agement approach incorporating direct localobservation of forage condition and changes, andrelating this to village-level or system-level graz-ing management, would be worth investigating

and may emerge from the ongoing livestocksystem studies.

Marketing and Enterprise Development

RelevanceMarketing and enterprise development are clearlyrelevant for a remote region in a diversifyingeconomy, and are consistent with national objec-tives. However, the objectives of this programcomponent need to be clarified and linked to theoverall mission statement. The directions takenappear to have been more opportunistic thanguided by a clear objective. The challenge is howto maximize leverage and be efficient in pick-ing potential winners. The program in this areahas been modified substantially over the yearsin response to the changing environment. TheAKRSP’s activities have ranged over a wide num-ber of enterprises. The challenges for enterprisedevelopment are formidable, including: “anunfavorable legal and regulatory environment,a lack of appropriate financial services, a lackof management technical and business skills,weak market information and business serv-ices, and poor infrastructure services” (AKRSP2000n).

The original marketing component concen-trated on improving the marketing of fruit (freshand dried), livestock, and grains, with an empha-sis on reducing losses of perishable products.Farmer training and credit were provided, butmuch of the activity was primed with grantfunding. The focus was on integrating market-ing to raise farm revenue by bulking up, by clus-tering neighboring villages in marketingassociations, and by grading, packaging, andtransport.

Early results were mixed, with few successesand some spectacular failures, especially wherethe AKRSP took on an agency function in mar-keting perishable produce. This led to empha-sis on processing to add value to local surplusproducts and reduce post-harvest losses, with anotable success in apricot drying. Apricots exhib-ited the high-value, low weight, local climaticcomparative advantage, high labor demand, andmodest technical demands that are needed forsuccess. Contract growing and air drying of apri-

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

1 2 0

cots, rather than sulfur drying, has demonstratedeven greater value added, and sustainability ofthis system needs to be sought by spinning offa sound enterprise.

From the mid-1990s the emphasis has movedtoward enterprise development, with the rangeof activities broadening away from the AKRSP’srural roots. Given the diversifying economy thiswas probably a relevant shift, but it has takenactivities into areas where the AKRSP has lesscomparative advantage. The latest strategy iscentered on providing additional business devel-opment services for small- and medium-scaleenterprises, mainly through partnership withexisting business service agencies and the pri-vate sector. In branching out from purely ruralproduction activities into small businesses, theEnterprise Development Section has supporteda wide range of enterprises, including vegetableseed production and marketing, shu (traditionalcloth) production and tailoring, apricot pro-cessing, input stores and shopkeeping, touristhotels, mining (marble, slate, stone, antimony,gems) and gemstone marketing, bottling, enter-prise promotion exhibits, and formation ofrelated business and professional associations.Notwithstanding this sectoral broadening, theAKRSP continues to respond to the special needsof NAC producers for improved marketing ofagricultural products, including fresh fruits, wal-nuts, peas, broiler chickens and small stock,potatoes, honey, and herbal teas. Emphasis onreducing losses of perishable products remainsa high priority, but, given past experience, thenote of caution in the Joint Review Missionreport of 1999 on minimal direct involvement inproduct marketing is worth remembering.

EfficacyMarketing and enterprise development to datehave shown modest efficacy—a tale of manyhopeful starts but fewer lasting results that sug-gests the need for some change in approach. Arelatively high level of failure for new enterprisesmay be difficult to avoid given the particular geo-graphic and logistic challenges characteristics ofthe NAC. Notable successes to date include apri-cot drying and packaging, which has greatlyadded to producer returns for this important

crop; seed potatoes, which started well but nowhas run into difficulties; vegetable seed pro-duction, which has grown to quite a large enter-prise with contract growers—but is still a“project” striving for profitability and privateownership; the shu fabric/clothing enterprise,which also needs to become private; and agri-cultural input supply shops, which are alreadyin private ownership.

EfficiencyIt is difficult to evaluate the efficiency of such adiverse collection of activities. No separate eco-nomic rate of return has been estimated for thiscomponent. Efficiency has probably been mod-est so far. It is to be expected that this compo-nent would need a startup subsidy as theAKRSP’s traditional strategy emphasizes exper-imentation, trial and error, starting small, admit-ting failures, and moving on if needed. Themarketing and enterprise development compo-nent has epitomized this approach.

Some individual marketing attempts havedemonstrated potential profitability, but have yetto be turned into viable, sustainable independ-ent operations (apricot drying and seeds pro-duction are the obvious examples). The supplyshops appear to be viable owner-operated busi-nesses. The new partnership approach, in asso-ciation with other business development agenciesand entrepreneurs, is promising and should pro-vide a more direct route to establishing viableenterprises, especially since the partners are likelyto be more selective in their choice of enter-prises and fill skill gaps in the AKRSP’s ranks.Within the AKDN there appears to be potentialsynergy to be had by the AKRSP’s enterprisedevelopment activities being more closely alliedwith and drawing on AKFED resources.

Institutional Development Impact and SustainabilityAs implied above, institutional developmentimpact through this component has been mod-est generally, but substantial in limited areas. Theearlier cluster marketing associations have tendedto wither once the AKRSP is no longer the driv-ing force. As more broadly for the program asa whole, the AKRSP faces its greatest enterprise

A n n e x e s

1 2 1

development challenge in devising ways toensure that its activities become sustainable ininstitutional terms. With skilled staff and ampleinitial resources, much can be achieved at theoutset of a new venture, but this is not devel-opment unless it is rapidly institutionalized,most probably as a profitable private enterpriseor in partnership with a corporate patron. Themore recent practice of working through part-nerships with existing agencies and private sec-tor entrepreneurs has a greater chance of leadingto establishing new, lasting institutions.

The AKRSP has also provided a wide rangeof relevant training in enterprise skills such asbusiness management, hotel management, book-keeping, post-harvest management, broiler farm-ing, fruit and vegetable processing, poultry feedand chick supply, embroidery, carpentry, automechanics, motor/generator rewinding, mining,gold panning, food processing, computer tech-nology, cobbling, and mineral appraisal. Thereis a need to evaluate the impact of this trainingon income generation in order to prioritizefuture training activities.

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

1 2 2

1 2 3

Draft Monitoring Mission Report(Baltistan Region), Stiles Assoc./CIDA,March 2001General: impressive results at the output andoutcome levels; institutionalizing of powerfulcommunity-based mechanism for participatorydevelopment; improvements a matter of finetuning; dynamic Baltistan environment withlocal elections, new village councils and electedunion councils, increased involvement of donorsand NGOs, and strengthened (but not enough)government line agencies. AKRSP meeting impli-cations of changes head on. AKRSP exemplifiesmany of CIDA’s key success factors for sustain-able results.

Social Organization: good progress—com-munity organizations effectively manage rangeof activities, VOs exceeded targets, but WOsbehind target (staff shortage) and greater effortneeded; impressively wide range of activitiesundertaken by VOs/WOs; and good progress indelivery of services/exceeding targets.

Women’s Program: reasonable progress inchallenging, conservative environment (but over-shadowed by VOs—exclusive domain of men);impeded by lack of coherent gender strategy,shortage of qualified female staff, and lack ofhome for women’s program and gender equal-ity in AKRSP (need champion at center); progresstoward solution is slow, and these activitiescould suffer if CIDA withdraws in 2003.

Capacity Development and Training: overallcapacity of Baltistan RPO enhanced; research,monitoring, and reporting generally exemplary,with staff nearly mastering CIDA’s results-basedmanagement, although reporting somewhat per-functory and reluctance to make needed changesto logframe agreement; high staff turnover in theMonitoring, Evaluation, and Research Section,

which could be compensated by consultants;some improvement in gender awareness couldhelp in improving the effectiveness of thewomen’s program.

Canadian Components: some aspects notintegrated with AKRSP or well understood, norreflected in logframe. [Other internal AKRSP/CIDA aspects not summarized.]

Mid-term Review Mission—Gilgit andChitral Regions, DFID, March 2000Sustainability and Future Directions: significantprogress on developing a strategy; prospectsfor significant future diversification of AKRSP’sfocus and resources, e.g., away from servicedelivery and toward closer relationship withgovernment; AKRSP has much to offer Pak-istan’s reform process and should re-orientplanning accordingly, particularly with respectto proposals for local elections and decentral-ization; suggest a “new look” AKRSP/DFID/Government partnership—e.g., “SustainableLivelihoods Development Programmes,” specif-ically to support the government’s “Local Gov-ernment Plan 2000.”

Chitral Sectarian Issues: (subject of a separatenote.)

Poverty: AKRSP has contributed to a sub-stantial reduction in poverty in NAC; infrastruc-ture and other schemes implemented bycommunities, especially irrigation, have hadmajor impacts and contributed to poverty reduc-tion; whole-community approach now needsto be replaced by targeted poverty reduction; andAKRSP to develop poverty-targeting criteria forinfrastructure projects and define poverty areas—for discussion with DFID.

Gender: DFID to review with AKRSP the rec-ommendations of upcoming consultancy.

ANNEX M: RECENT DONOR PROGRAM EVALUATIONS OF THE AKRSP: A SUMMARY OF MAIN CONCLUSIONS

Enterprise Development: need for ED strategythat is being developed for mainstreamingthroughout AKRSP, of which key componentwill be staff training and exposure visits.

Credit and Savings: known for strong focuson financial sustainability, but credit volume isnow declining/reasons being investigated; AKRSPplaying key role in work on proposed microfi-nance bank.

Environmental Screening: noted successfulintroduction of environmental screening forinfrastructure projects, and should be applied toall projects; and follow-up workshop agreed.

Impact Assessment: impressed with high qual-ity of work, demonstrating major AKRSP impact,despite attribution problems; and OED shouldbe invited to evaluate program for fourth time.

Increased Funds for Infrastructure: agreedthat increased funding could be justified, andwould be discussed in context of evolving sus-tainability strategy.

Natural Resource Management: significantpart of program but expected results not deliv-ered; new NRM strategy being prepared in timefor next AKRSP funding proposal.

Budget: funds expected to be spent by March2003; consultancy funds well used, especially ontraining.

Extension of AKRSP/Gilgit: agreed for oneyear.

Networking with other RSPs and NGOs: notedsubstantial impact of AKRSP on other RSPs inPakistan and in the South Asia region; endorseAKRSP networking with other RSPs and NGOs;and AKRSP has key role in rural developmentforums in country and internationally.

Lesson Learning: AKRSP to revise its lessonlearning and dissemination strategy and consultwith DFID.

Joint Review Mission, 1999 (all donors)Social Organization: highly successful activitiesresulting in significant income increases; mustbalance new initiatives with continuing nurtureof VOs/WOs; new support packages needed(PPIs) to maintain institutional viability, but withself/non-program finance; women’s packagesneeded to prepare them for leadership; AKRSPneeds more female senior staff; and caution is

in designing special poverty projects regardingtraditional relief systems.

Credit/Enterprise Development: AKRSP pro-vides unique service; caution on impact studies(criteria suggested); modest poverty alleviationpotential of credit; credit/enterprise opportuni-ties for women are modest; balance credit deliv-ery objectives and financial sustainability;suggestions for revitalizing savings; EDC to berealistic on SME capabilities (given staff con-straints); need to identify agricultural marketingopportunities and devise more effective deliv-ery of business services.

Natural Resource Management (NRM): poor-est families heavily dependent on farm income;recommend analysis of poverty aspects of NRMsubsectors; study land tenure issues; classifydiverse farming systems (before focusing onany one zone such as single cropping); empha-sis on food and fodder security (through par-ticipatory research/strengthened links withresearch institutions), and training master train-ers; need coherent subsidy policy (end input sub-sidies); continue land development projects;research outputs to guide on possible AKRSPassistance for management of common propertyresources; cautious/minimal involvement withproduce marketing; continue extension for cashcrops, and applied research on high-value non-perishable cash crops; studies needed on effectsof wheat subsidies, poultry-women-povertynexus, and market for trophy hunting. NRMvision is moving to training master trainers, pro-viding market information, facilitating traderlinks, and ensuring input quality.

External Involvement and Interactions: exter-nal perceptions consistently positive; develop-ment model acknowledged as important, butAKRSP should be open to learning from otherdevelopment initiatives; surprising number ofexternal relations add to effectiveness of AKRSP;good government relations, which somewhatimproves government capacity; more effortrequired with NWFP government; synergiesfrom cooperation with other AKDN agencies, butmixed opinions on need for formal cooperationmechanism; AKRSP always non-sectarian, butAKDN less so, and this causes confusion (anAKDN issue); AKRSP active on national stage

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

1 2 4

regarding microfinance, but less on RSP forumsto which it should respond positively; shorterannual report is considerable improvement.

Sustainability: protracted debate; some AKRSPfunctions needed in NAC for 20 years; endow-ment concept raises issues discussed; AKRSPshould not split into regional RSPs.

Impact: highly positive, well documented,and responsive to expressed needs; burden ofassessments falls on overworked field staff, whoshould be consulted on realistic indicators andmethodologies; new initiatives need to be awareof potential impacts on established values andinstitutions; and there may be need for aug-mented social organization field staffing andtraining in program impact assessment.

Afterword: program should continue to focuson economic and institutional development ofNAC, principally through VOs and WOs; AKRSP’scredibility arises from its commitment, continu-ity of procedures, and the terms of partnershipswith communities; and the long-term future ofNAC depends on unconventional vision, whichAKRSP should stimulate.

Annual Review Mission (Gilgit andChitral Regions), DFID, 1999Strategic Approach: endorses AKRSP’s strategicplanning (the sustainability and NRM strategyreviews) and recommends AKRSP catalyze widerregional strategic planning.

Gender: good progress, but better focusneeded to involve women more in planning,broaden their options, meet their desire for safedrinking water, and avoid increased commer-cialization that marginalizes women; recom-mends increasing AKRSP female field staff andWO contacts, improve gender analysis, disag-gregate gender data in all activities and dis-seminate, expand monitoring of women’s rolein all program activities, increase women’s activ-ities in non-Ismaili areas, and set a target datefor universal access to safe drinking water.

Equity: good progress in assessing equity/poverty impacts and in developing criteria foridentifying resource-poor communities; pilotpoverty projects being developed by Gilgit RPO;recommends sample surveys, data sharingaround program, mapping of poor areas, mon-

itor benefit distribution, and make communitiesaware of equity; monitor V/WO representative-ness, use equity analytical tools in planning,work on poverty targeting and related method-ologies, and continue to increase support fornon-Ismaili villages and monitor programresource allocations.

Infrastructure: AKRSP very successful indeveloping schemes and encouraging commu-nity execution and maintenance, bringing con-siderable livelihood benefits; recommendsfinding ways to enhance impact of infrastructureprogram (sample surveys), improve prioritizationof schemes with economic analysis (by con-sultant), review staffing needs of MIES and takeon women social organizers for sanitation/hygiene schemes, review subsidy levels for irri-gation and microhydel schemes, develop jointtechnical training with SAP and train governmentstaff in social organizer work, provide technicalassistance for AKDN agencies in water sup-ply/hygiene technologies, and organize a Moun-tain Technology Workshop.

Natural Resources Management: takes 20 per-cent of DFID funds for diverse activities, whichmay not all be justified by impacts, and roles notclear of government, private sector, and NGOsin NRM activities; recommends refocus onsmaller number of viable activities in frame-work of integrated farming systems approach,and implement a participatory research anddevelopment approach to identify priorities andtechnologies.

Environment: most activities are environ-mentally beneficial but greater consideration ofimpacts is needed; recommends introducing(mostly simple) screening of all projects (espe-cially an EIA of Risht coal mine), staff trainingin impact assessment, and hiring of consultantsto assist institutionalizing the process and to doan EIA of Risht coal mine.

Enterprise Development: good progress inpromoting job creation and skills upgrading ina generally difficult but nascent enterprise cul-ture; recommends (in addition to some internalprocess details) resolving staffing requirementsof EDS and EDC, deciding EDC approach toenterprise finance, develop local staff to take ontasks that are currently contracted out, conduct

A n n e x e s

1 2 5

impact assessments of all EDC/EDS schemes,send staff for international training.

Savings and Credit: very impressive systemstreamlined to two products—collateralizedgroup loans and uncollateralized small individ-ual loans, very good credit performance andsound management systems; recommends useof logframe indicators, developing impact assess-ment instruments, and staff networking withorganizations inside and outside Pakistan, andattendance at international conferences/train-ing courses.

Other Recommendations (aside from internalDFID process needs): survey importance ofmigrant labor and impact on women and chil-dren at home; continue to develop links with andseek to influence government policies; inde-pendent evaluation of the Training and Learn-ing Programme (social analysis) in 2000; variousconsulting inputs; need to develop a “commu-nications strategy” to disseminate AKRSP learn-ing; AKRSP to review training program andencourage inclusion of government staff; goodprogress on impact assessments but AKRSP mayneed to allow MER greater flexibility in hiringconsultants; AKRSP to discuss with donorswhether a fourth OED evaluation is required; andworkshop for early 2000 to review logframe forreporting to DFID.

From Small Farmer Development toSustainable Livelihoods—A Case Studyon the Evolution of AKRSP in NorthernPakistan, Khaleel A. Tetlay andMuhammad A. Raza, Aga KhanFoundation (Pakistan), November 1998

Major Lessons and Emerging Challenges:Principles and Practices of Management: utilityshown of a “participatory, flexible, and catalytic”approach; rural people can undertake develop-ment with a trustworthy partner and will adoptnew forms of organization and keep partnershipto gain direct benefits; and lists 13 points of(good) management practice from first OEDevaluation of AKRSP.

Changing Socio-economic Contours: greatsocial changes since program was planned, inresponse to government initiatives and assis-

tance from NGOs; and pace of socio-economicdifferentiation is accelerating.

Moving Beyond the VO: input delivery byAKRSP to VO was major intervention at expenseof attending to other social arrangements, butnew strategy represents a more pluralisticapproach to include social targeting and moreof the social development agenda.

Women’s Real Agenda: going beyond poul-try and vegetable schemes with WOs, new chal-lenge is to help women meet other strategicneeds.

Village-level Infrastructure Development: afterPPIs have met all criteria—productivity,sustainability, equity, and environmentallysound—challenge is to articulate key prioritiesfor future, including access for others to spon-sor infrastructure.

Natural Resource Management: need to break15-year tradition of package delivery and train-ing, not substitute for government extensionservice, and identify where concentrated effortis needed, such as common property resourcemanagement; potential for linkages with nationaland external research institutions; suggest atten-tion to non-land-based agricultural services enter-prises such as seed project and Shubinak; andchallenge of initiating work on “low externalinput” agriculture, particularly in high-altitudeareas.

Credit Program: preparations proceeding forturning the donor-supported revolving creditfund into a development finance institution; andchallenge of developing sufficient revenue forfinancial sustainability.

Enterprise Development: importance of non-farm activities is increasing; and need for larger-scale enterprise is rationale for proposed“Enterprise Support Company”—challenge oflaying strategic and organizational groundwork.

Revisiting Human Resource Development Insti-tute: did not work out in late 1980s, but needto consider whether to reopen that debate—partof AKRSP or of AKDN, what niche areas, forwhom, manpower and cost recovery?

Conclusion: over 15 years AKRSP has “blazeda path of rural development”; being replicatedwithin and outside Pakistan; ways of earninglivelihoods have changed, as have people’s

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

1 2 6

expectations for future and their role in deter-mining that future; and many second-generationchallenges and opportunities for AKRSP, whichits record suggests it will surmount to reach ahigher plateau.

Joint Monitoring Mission, 1997 (sixthannual mission for the donorsconcentrating on Baltistan and Chitralsub-programs, and the Gilgit core office)General: 1997 saw a leaner and stronger, morefield-oriented AKRSP emerge from the transitionof 1996/97; seeking a balance between equityand growth; improved annual planning; startingon new directions for this funding phase.

Credit and Savings: impressive improvementsin professionalization; systems strengthened;segregation continues of credit and savings oper-ations in anticipation of creating a bank (nowdelayed); savings moderated, while credit stillgrows; swing from VO/WO credit toward enter-prise loans; emerging problems in Baltistan andwith LDO lending; and credit and savings needsto become more self-sufficient.

Enterprise Development: has become morefacilitating than directly involved in enterprises;mixed results with new corporate credit (whichshould be halted); activities depend on a fewpeople; need staff strengthening and manage-ment processes in preparation for creating anEnterprise Support Company.

Natural Resource Management: program con-tinues to plan NR development and build insti-tutions for NRM; creation of Begusht CommunityGame Reserve has strengthened sense of own-ership of commons; agriculture, livestock, andforestry undergoing increasing commercializa-tion, and AKRSP needs to assess impact of thistrend on NRM strategies; AKRSP has begun anintegrated approach to breed improvement andfodder security for remoter areas where livestockis the main livelihood.

Agricultural Production and Marketing: sen-sible focus on horticulture, vegetables, and pri-vate sector initiatives; constraints to productiongrowth are the (harsh) environment, logistics, andthe subsidy on government wheat sales; with highmale labor migration, much of the burden, andmany of the benefits, of agriculture fall to women;

and two cautions—AKRSP resource constraintsrestrict its role (cannot be a surrogate for gov-ernment agriculture and research services), andhorticultural development depends on reliableand profitable post-harvest arrangements.

Mountain Infrastructure and Engineering:most tangible impact/impressive performance;Chitral hydel revolution continues; a noticeableimprovement in quality of project proposals;but larger cluster schemes require more complexoversight and careful review of all documenta-tion to avoid cost overruns.

Gender Issues: commercialization and devel-opment of interest groups/LDOs in danger ofleaving women further behind; economicempowerment needs to be accompanied bysocial empowerment; WOs beginning to take upwater supply projects; WSO staff not increasingwith WO growth; AKRSP places women’s par-ticipation and benefits high on agenda and allstaff should be held accountable for meetingthese goals.

Social Development: good senior-level dis-cussions but little attention given in field toinstitutional development of community organ-izations; social development course being pre-pared; attention being given to equity as well asbroad-based growth, including poor areas andpoor groups in villages; policy paper needed onpoverty strategy; and participatory implemen-tation and monitoring should be developed.

Training and Human Resource Development:impressive level of community training contin-ues; staff training enhanced with external assis-tance; personnel management practices beingstrengthened; enthusiastic TSU staff need furthersupport and guidance; and need to build andsupport some staffing areas (social over techni-cal staff).

Linkages: Joint Monitoring Mission conductedcomprehensive review of linkages; many oppor-tunities, but caution in order—government link-ages constrained by declining resources, anddonors’ expectations for NGOs to solve gov-ernment capacity problems should be refocusedon donor country strategies, such as civil serv-ice reform; and linkages with other AKDN agen-cies are satisfactory, particularly with the healthservice.

A n n e x e s

1 2 7

Management: need for stronger core capac-ity in social development, gender, and enter-prise development; also need greater regionalcapacity in social organization (and gender inBaltistan); reorganization and other managementchanges have made it possible to improve fieldprograms, and RMT can appraise proposals,logframe permits verifying against objectives(should be expanded from Gilgit to program-wide), and more can be done to include strat-egy in planning; and internal auditor still needed.

Exit Strategy: donor-driven demand for an“exit strategy”; term seems abrupt and longer-term support is warranted; viability of an endow-ment to ensure financial security should beinvestigated.

The Future of Joint Monitoring: if donors con-tinue Joint Monitoring Mission in 1998 it shoulddrop linkages topic, cover all regions by split-ting, but less comprehensively, and spend threeweeks in field.

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

1 2 8

Chapter 11. Defined as:Relevance—The extent to which the objectives

of the program are consistent with the currentdevelopment priorities of the country and theNorthern Areas and Chitral (NAC).

Efficacy—The extent to which the objectivesof the program were achieved, taking intoaccount the relative importance of the differentobjectives.

Efficiency—The extent to which the programachieved a return higher than the opportunitycost of capital and achieved benefits at the leastcost compared to alternatives.

Institutional development impact—The extentto which the program improved the ability of theNAC to make more efficient, equitable, and sus-tainable use of its human, financial, and natu-ral resources.

Sustainability—The resilience to risk of netbenefit flows from the program over time.

2. AKF, “Proposal for a Rural DevelopmentProgram in the Northern Areas of Pakistan,”November 27, 1981.

Chapter 21. The poverty goal called for cutting in half

the proportion of people living in extremeincome poverty of less than US$1 a day by 2015.Currently, average per capita income in theNorthern Areas is about 50 cents US per day.

2. Poverty incidence is the percentage ofpeople who fall below the current Pakistanbasic needs level, the measure of which is thecost of achieving a minimum bundle of basicneeds at current prices.

3. Pakistan has embarked on a massive pro-gram of decentralization by requiring electionsto a series of local bodies called Dehi Councils,with a District Council at the apex. This programwas scheduled for completion in 2001. In theNorthern Areas, however, the first stage involvedestablishing village-level councils through aprocess of nomination and acclamation. Bodieswith more democratic processes are expected tofollow, but the timetable is uncertain. See Pak-istan Government, National ReconstructionBureau 2000. For an account of AKRSP’s role,see Akram 2001, p. 3.

4. Government outlays in the Northern Areason agriculture, animal husbandry, forests, fish-eries, local bodies, and rural development totaledapproximately Rs. 204 million in fiscal 1999/2000,while the AKRSP’s total expenditure was aboutRs. 199 million. However, the proportion spenton operations as opposed to overhead wasabout 50 percent in the AKRSP and less than 5percent in government.

5. In fiscal 1999/2000 total government out-lays in the Northern Areas were Rs. 2,546 mil-lion, of which Rs. 910 million was fordevelopment. For fiscal 2000/2001 the devel-opment budget in Chitral is Rs. 35.8 million.Details of the nondevelopment budget are notavailable.

6. For example, in Chitral District the AnimalHusbandry Department has no operational fundsin its budget for 2000/2001. A senior districtagricultural official explained how he must seekformal permission from the commissioner’s officeto leave his station in town.

7. As explained to the evaluation team by Mr.Sang-e-Marjan, Chief Secretary, Northern Areas,in a meeting on May 10, 2001.

8. The mission reviewed the coefficients usedand obtained independent comments on theNRM coefficients from two specialists who hadworked in the area before. The mission con-cluded that there was over-optimism in someenterprises, but that it was not excessive. Themain coefficients where adjustments wouldlower benefits were wheat seed, which showedrather high yield increases in the absence of fer-tilizer, and wheat straw, where differencesbetween improved and local varieties weresubstantial.

9. AKRSP 2000c; Andrew Foster (consultant),unpublished mission working note.

10. Microhydel is the common term in Pak-istan. These are small-scale, usually village-sizedcapacity, hydropower generation units that takewater from small channels run from a snowmeltwater source and run it down a tube to a gen-erating turbine at the bottom.

11. Economists normally use a fourfold clas-sification of public and private goods: privategoods, such as land, houses; toll goods, such astoll roads and bridges; open-access or common

1 2 9

ENDNOTES

property goods, such as grazing land, someforests, fisheries; and pure public goods, suchas air, information, a clean environment.

12. The Joint Review Mission’s Final Reportcontains a useful if largely theoretical annex onsubsidies. But it also concludes that the AKRSPshould formulate a well thought through subsidypolicy.

13. A forum consisting of the ADKN agencygeneral managers meets occasionally and theboards have some common members.

14. “Collaboration with the AKRSP at theoperational level has strengthened governmentcapacity in a lasting way to a very limited extent.”Joint Review Mission 1999, p. 77.

15. The AKRSP salary package (salary andbenefits) is about the same as the government’s,but there are half as many grades and no stepswithin grades. Thus, in the AKRSP, salaries aremore flexible and promotion is faster.

16. In particular, the universities of Bath andEast Anglia in England.

17. The AKRSP’s stated gender policy is, how-ever, clear and progressive. See Gender Policy,AKRSP 2000a.

18. A recent gender assessment of AKRSPprograms in Baltistan reported a corporate cul-ture of discomfort about gender relations andissues. The same report also highlighted thegap between the message the AKRSP is pro-moting in the community and the structure andpractices within the AKRSP. See Moffat 2001.

19. A recent report by a gender consultantmakes the useful suggestion that an existingmale member of the core team should be nom-inated to share this role. See Seeley 2000.

20. The NRSP has the greatest geographic cov-erage among Pakistan’s rural support programs.It works in more than 30 districts and has anannual expenditure of more than Rs. 200 million.It organizes socially viable groups of 20 to 30households and helps them identify micro-investment plans and opportunities at the house-hold, group, and village levels. Plans andschemes are implemented by individuals andcommunities through their own efforts and withcredit provided by the Khushhali Bank.

21. This is the outcome of a recent decisionby the Provincial Government of the Punjab to

remodel its entire rural development programalong NRSP lines.

22. A fund of $10 million yielding 10 percentwould cover about 7 percent of the AKRSP’soperating costs, Sustainability and the AKRSP:Plans for the Future, June 2000, p. 34. Theendowment idea has been on the table for someyears. The 1997 Joint Monitoring Mission reportrecommended the investigation of the viabilityof an endowment.

23. An analogy with agricultural research isapposite.

Chapter 31. The evaluation team found the 1998 “Insti-

tutional Maturity Report” a useful analysis in somerespects, but with three weaknesses. First, it cat-egorized community organizations based on indi-cators, but did not attempt to account fordifferences. Second, the indicators were selectedby the AKRSP and not by villagers. The AKRSP’sunderstanding of “maturity” could be at variancewith villager perceptions. Nor were the indicatorschosen obviously reflective of maturity. For exam-ple, does the acceptance of a subsidized NRMpackage really indicate institutional maturity, orjust being in the right place at the right time tosay yes? In summary, the analysis did not explainwhy some organizations remain vigorous whileothers have declined. Third, the report did notdraw sufficiently from the evidence of the appar-ent modest correlation between institutional matu-rity and number of years under the program.

2. DFID’s Annual Review Mission in 1999pointed to a special area that the AKRSP mayhave overlooked in working with WOs when itsuggested that a survey be done of the impor-tance of male labor migration and the impact onwomen and children left behind. There mayalso be a linkage here with the special creditneeds of women that the new bank couldexplore.

3. The 2000 DFID Mid-term Review Missionreport noted that whereas the “whole-commu-nity approach” had made a major impact, thetime had come to replace it with a targetedpoverty reduction approach.

4. The genesis of a more targeted povertyapproach has been slow but thorough. The 1997

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

1 3 0

Joint Monitoring Mission report noted that atten-tion was being given to equity as well as broad-based growth, including poor areas and poorgroups in villages, and that a policy paper wasneeded on poverty strategy.

5. Experience elsewhere has identified someways to handle social exclusion. The first step isto identify non-participating households. This canbe done by a survey of all households to recordthe activities of men and women (separately).This is followed up with an investigation, usuallyby co-villagers, to find out why non-participatinghouseholds are not participating. Usually the rea-son is that they are, or feel they are, lower statusor poorer than other households, or unwanted bythem. Once the reasons are clear, established vil-lage groups can be encouraged to look for waysto solve the problem. This is what is now hap-pening in the AKRSP’s poverty project. The prob-lem is that the poverty initiative is coming at a timewhen the AKRSP is already moving away fromintense village involvement, and, in any case, itoperates in only a handful of villages.

6. This section draws on women’s perceptionswhere possible, and also on the useful impactreports on aspects of the women’s program pre-pared by the Monitoring, Evaluation, and Researchand Field Management Units in each region.

7. The CIDA Monitoring Mission Report ofMarch 2001 suggests that the Baltistan women’sprogram is constrained by lack of a coherent gen-der strategy, shortage of qualified female staff,lack of a home for the women’s program, andgender equity in the AKRSP. The DFID AnnualReview Mission in 1999 (on Gilgit and Chitralregions) called for increasing female field staffand the level of contacts with WOs, particularlyin non-Ismaili areas.

8. The AKRSP tables do not segregate train-ing for WO members, but, for example, poultrytrainees alone (all women) total 4,000 over thefull period.

9. It aims to equip villagers with the skills toalter substantially the situation of the poorest.Poverty Committees, comprising both men andwomen, are formed to help identify the poorestand then work with them to develop and imple-ment individual poverty reduction programs.They are given an endowment fund for this pur-

pose. Results so far are encouraging, at least inone of the two villages visited. The vast major-ity of targeted households have substantiallyincreased their incomes as a result of the program;they have also joined the community organiza-tion, although the outcome of this is not known.

Chapter 41. Other measures under consideration to

reduce risks of internal lending include addingtraining of community organization office-bearers in internal lending management andadding internal lending audit procedures toaudits of community organizations done by vil-lage accountants.

2. Includes loans overdue for any period.The loan loss ratio is the total write-off dividedby the average loan outstanding.

3. Total amount of loans with balances inarrears for which the AKRSP does not hold anycollateral.

4. Does not include financial costs.5. The NRSP comes closest to the AKRSP in

terms of the maturity of its Microfinance Program,but the Microfinance Group’s Performance Indi-cators Report for 2000 does not give figures forthe NRSP.

6. A network of NGOs that disseminates bestpractice and promotes microfinance in Pakistan.

7. DFID 1998. The rapid growth of schemesin Chitral is attributed to the high cost and unre-liability of public electricity supplies there, com-pared with subsidized supplies in Gilgit andBaltistan. This poses a public policy issue withrespect to the disincentive for private microhy-del schemes in Gilgit and Baltistan.

8. Technical assistance has been provided inPakistan to the IFAD-funded Chitral Agricul-tural Development Project, to schemes in Dir-Kohistan (IUCN), Patan-Kohistan (KODEP),North Waziristan Agency, and Murree (IUCN),and in Afghanistan, as well as to private micro-hydel schemes, and technical assistance andtraining has been provided to the InternationalCenter for Integrated Mountain Development(ICIMOD), Nepal, and to staff of two Afghanistanaid agencies.

9. With respect to comparative costs across theNAC, an AKRSP study compared 230 projects

E n d n o t e s

1 3 1

constructed in 1998/99 by regions. The mainfindings were: costs per household were sig-nificantly lower in the Gilgit region becauseschemes had more families; Chitral had signifi-cantly higher unit costs per acre and per foot ofchannel because of limited land; communitycontributions to costs were from just over 20 per-cent in Baltistan and Chitral, to as much as 43percent in the Gilgit region, reflecting ability topay; and community contributions were highestfor link roads—the predominant public goodcomponent of infrastructure interventions—whereas equity and efficiency would suggest thatpublic goods should get the largest subsidy.DFID’s Annual Review Mission of 1999 recom-mended reviewing subsidy levels for irrigationand microhydel schemes.

10. AKRSP 2000a, p. 19, section 4.11. The common view has been that wheat

production has no comparative advantage inthe NAC. However, the issue is complex. Apartfrom the fact that there is a modest quality dif-ference with wheat grain to some consumers (notso modest in some cases—in Teru, due to bothquality of a high-altitude Tibetan variety andtransport, the price was substantially above thesubsidized price of wheat from down-country),wheat is also a fodder crop. About two-thirds oflivestock nutrients come from cereal straw. Theyield of fodder from a local variety can be morethan twice that from an improved variety. More-over, winter wheat can substantially increasefodder production if uncontrolled livestock graz-ing can be managed. Also, increasingly, sea-sonal labor requirements will be a factor, aswill competing demand for more high-valuecrops. It is therefore not so obvious in a systemscontext what would happen to wheat produc-tion if wheat from down-country were not sub-sidized. This is an issue warranting some farmmanagement research that incorporates risk.The 1999 Joint Review Mission report called forstudies of the effects of the wheat subsidy.

12. The subcomponents—including partici-patory variety selection, seed supply, livestockhusbandry support, veterinary supplies, breedimprovement for large stock, poultry, and mas-ter and village specialist training—are addressedin Annex L.

13. In 1997, the constituents of farm incomeswere as follows: crops/vegetables, 34 percent;fruit, 12 percent; forestry, 14 percent; and live-stock, 40 percent.

14. The 1997 Joint Monitoring Mission reportnoted that the AKRSP had begun an integratedapproach to breed improvement and foddersecurity for the more remote areas, where live-stock are the main livelihood.

15. In the context of poverty analysis, the1999 Joint Review Mission suggested classify-ing the diverse farming systems before focus-ing on one zone (such as single-cropping),and DFID’s Annual Review Mission of 1999recommends refocusing NRM activities on asmaller number of activities in a framework ofan integrated farming systems approach. Ear-lier, Whiteman (1985), in an exceptional andarguably underutilized piece of work, laid thetechnical groundwork for farming systems clas-sification, but without the social and economicoverlay.

16. These are areas of dry, rocky rangelandextending up to the high mountain pasturesabove the irrigation channels. The area is ahigh-altitude desert with limited forage growthfrom direct rainfall. But there is a summer flushwatered by glacier and snowmelt providing sea-sonal pasture—the source of most of the annuallivestock weight gain.

17. A social accounting matrix, or SAM, explic-itly draws out forward and backward linkagesin an economy, including a regional economy,and shows how value added generated in theeconomy is distributed across the principal eco-nomic agents—businesses, households, and thestate. A full SAM should not be estimated for theNAC—there is neither time nor data with whichto do so. But the framework of a SAM used asan organizing device would provide a logical andconsistent appraisal of where the enterprisedevelopment program should focus.

18. This external income aspect was not cov-ered in the February AKRSP strategy discussionpaper.

Chapter 51. The AKRSP is actively considering the

options. See, for example, AKRSP 2001.

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

1 3 2

2. The challenge in building governmentdevelopment capacity lies in transferring theconcepts and tools used by the AKRSP in itscommunity work to the key branches of gov-ernment present in the rural areas in ways thatsupport and reinforce the democratic and admin-istrative reforms at the core of the government’sdecentralization strategy. Failure to do so, therebyallowing other agencies to continually substitutefor the state, leaves unresolved the fundamen-tal problems of accountability and governancein the relationship between citizen and state.These issues are a commonplace in the devel-opment literature. For a discussion in relation tothe AKRSP, see Wood 1996, p. 15.

3. The broadening of mandates among theAKDN institutions to include all peoples in theNAC (rather than the Ismaili community only)allows the AKDN to look at the developmentchallenge in much the same way as government.

4. Dehi Councils have been charged withsubstantial duties, including the managementand supervision of all government staff andassets within the boundaries of the village.However, it is not clear where the requisiteresources will come from. See Ordinary Gazetteof Pakistan, Part II, Islamabad, 26th November1999.

5. The DFID Mid-Term Review Mission ofMarch 2000 saw prospects for significant futurediversification of the AKRSP’s focus andresources—for example, away from service deliv-ery and toward a closer relationship withgovernment.

6. These challenges are not new. All are artic-ulated, perhaps in a lower key and with differ-ent emphasis, in Task Force on Sustainability2000. For example, p. 21 gives evidence of pastcollaboration with government and the need tohelp strengthen the public sector to play its des-ignated role, while p. 22 outlines the need forcloser integration and coordinated strategic plan-ning in the AKDN.

7. An examination of possible closer inte-gration in the AKDN and the problems involved,commissioned by the Aga Khan Foundation, isalready under way.

8. Ten years at least, but more likely 20 (seeJoint Review Mission 1999, p. 89). In effect, the

partnership would be a development coalition.9. Such concerns are not new. The 1997 Joint

Monitoring Mission report reviewed linkagesand noted that, while there are many opportu-nities, caution was also in order as governmentlinkages were constrained by declining resourcesand the AKRSP could not be expected to solvegovernment capacity problems.

10. “Government departments in the main re-main under-resourced and under-motivated.”The Task Force on Sustainability 2000, p. 10.

11. The apparent success of the NRSP can beattributed, in part at least, to the strength of itspartnerships, especially with government. But,as has been noted in earlier OED reports, rela-tions that are too close risk co-option.

12. However, with a development coalition,the AKF would need to consider playing a lessdirective role than at present.

13. In this way, donors could play a direct partin the development consortium. However, allparties, including government, should contributeto the fund, proceeds from which would onlybe used for partnership purposes.

14. A functions-based approach determinesthe appropriateness of a community organiza-tion based on the functional outcome sought. Itaccommodates the possibility of continued sup-port for multipurpose VOs or WOs where thatis the articulated village need, but it also accom-modates more specialized organizations suchas production or marketing organizations wherethat function is given highest priority by a com-munity group.

15. In developing a differentiated approach,demand for services or inputs should not be con-fused with their relevance to the poor. Demandis not the same as need. Demand says moreabout awareness of an individual of what is onoffer, incentives and subsidies, and knowledgeof the rules of access. Meeting “demand” is notabout equity, although it may offer efficiencygains from an organizational point of view. Notresponding to demand may be a key part of adifferentiation strategy. The AKRSP needs to beclear that there are tradeoffs in responding todemand.

16. Most of which, one should remember,make charitable contributions to Ismaili funds.

E n d n o t e s

1 3 3

17. This would call for a time-bound, objectives-based agreement between the AKRSP and everyorganization, with clear obligations on eachside. Independence becomes a joint objective.Every participating group has a plan. From theAKRSP side, it involves defining inputs and indi-cators for success. It ensures efficient use of stafftime. Visits have an objective related to the group’sability to become independent. A withdrawalstrategy is particularly helpful for new groups. Ifindependence is the objective from the outset, andinputs are presented as time-bound, then efficientand innovative ways of moving up the institutionalmaturity ladder are more likely to be tried. Flex-ibility would be required, with different rules forservice delivery operating for groups at variousstages. This approach will work less well for vil-lage institutions in isolated or resource-poor areasfor which the AKRSP is likely to be the soleprovider. In these circumstances the incentive tooperate independently may be low.

18. AKRSP India, for example, is movingtoward federations (8 to 30 organizations per fed-eration). Benefits there include shared goals,economies of scale in input supply and market-ing, stronger negotiation with external agencies,legal recognition for village organization mem-bers, improved handling of inter-village conflict,support for weaker village organizations, address-ing needs not supported by the AKRSP (India),and promoting new village organizations with lessAKRSP support. Risks include economic powershifting away from the village organization—addressed by ensuring that organizations canwithhold dues—capture by political parties, andflooding the federation with too many activities.

19. The evaluation team found a differenceof perception between the Northern Area gov-ernment and the AKRSP on the role of the DehiCouncils. The government expectation seems tobe that they will, in due course, handle allinvestment decisions at the village level. TheAKRSP expectation seems to be that they willhandle only government-funded investments.What happens in the end will probably dependmore on village attitudes and community evo-lution than on government or the AKRSP.

20. The 1999 donors Joint Review Mission sug-gested that women’s packages be developed

specifically as a means of preparing women forleadership.

Annex E1. The survey contained 39 questions, repro-

duced in this annex in full. The response ratewas 76 percent (170 out of 224), with little vari-ation by grade, gender, or location.

2. In general the lowest scores were given bystaff in the Core Office, who interacted with theteam more than most others and who are morefamiliar with management tools and analyticalprocesses. This lends credence to the upwardbias hypothesis. Women gave uniformly higherscores than men. Whether this is a more severeexpression of upward bias or a genuine posi-tivism is impossible to discern.

3. There was only one opportunity to explainthe aims and procedures of the survey to AKRSPmanagers and no opportunity to address staff atlower levels. Thus, not only was this surveynovel for AKRSP staff, it was also a surprise.

Annex L1. Other measures to reduce risks of internal

lending under consideration include addingtraining of community organization office bear-ers in internal lending management and addinginternal lending audit procedures to villageaccountants’ audits of community organizations.

2. Includes loans overdue for any period. 3. The loan loss ratio is the total write-off

divided by the average loan outstanding. 4. Total amount of loans with balances in

arrears for which the AKRSP does not hold anycollateral.

5. Does not include financial costs.6. The NRSP comes closest to the AKRSP in

the maturity of its Microfinance Program, but theMicrofinance Group’s Performance IndicatorsReport for 2000 (AKRSP 2000h) does not give fig-ures for NRSP.

7. Skills such as in construction of roads insteep and unstable terrain, identifying viablewater sources, channel alignment and con-struction, rock blasting, stonework, and bridgedesign and construction.

8. Technical assistance has been provided inPakistan to schemes in Dir-Kohistan (IUCN),

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

1 3 4

Patan-Kohistan (KODEP), North WaziristanAgency, and Murree (IUCN), and in Afghanistan,and technical assistance and training has beenprovided to the International Centre for Inte-grated Mountain Development (ICIMOD), Nepal,and to staff of two Afghanistan aid agencies.

9. Defined by the Development AssistanceCommittee as “The totality of positive and neg-ative, primary and secondary effects producedby a development intervention, directly or indi-rectly, intended or unintended” (DAC/OECD2001).

10. A consultant study of 132 (100 AKRSP and32 non-AKRSP) infrastructure projects by theMountain Infrastructure and Engineering Sec-tion (Kwaja 1999), which supports other find-ings, concluded that infrastructure projectswere in a reasonably good physical state,although there were some weaknesses in main-tenance; that extensions of older village proj-ects exhibited better maintenance than newprojects started from scratch; that new projectsstarted from scratch exhibited higher benefitsand more equitable distribution of benefits;that the level of maintenance of AKRSP proj-ects was similar to other external agency proj-ects; that a majority of projects had substantialbenefits but that the range across villages isquite wide; that the distribution of benefits isgenerally very equitable; and that there seemsto be a trend in recent years toward projectswith more equitable benefits.

11. A Regional Comparative Cost Analysis ofAKRSP’s Investment in Infrastructure Projects(AKRSP 2000g). The scale of village inputs isbelieved by staff to be higher than is often cred-ited for three reasons. First, infrastructure “costs”quoted in AKRSP documents frequently do notinclude village contributions, thus understatingthe level of infrastructure investment achievedby the program. Second, the estimated 30 per-cent contribution from the villages to the cost ofinfrastructure schemes is the “book cost” esti-mated by the AKRSP. AKRSP staff have rou-tinely noted that the hours of village labor andquantities of local materials provided by the vil-lage organization usually amount to a valuewell above that formalized in the agreement—typically double the estimate. Moreover, AKRSP

staff say the grant amounts to not more than athird of the public sector funds needed for con-struction by contractors of similar works and inthe absence of a participatory approach, the sit-ing, quality of construction, and sustainability ofpublic works has often proved to be problem-atic in the Northern Areas.

12. The common view has been that wheatproduction has no comparative advantage in theNAC. However, the issue is complex. Apartfrom the fact that there is a modest quality dif-ference with wheat grain to some consumers(not so modest in some cases—in Teru, becauseof both the quality of a high-altitude Tibetan vari-ety and transport, the price was substantiallyabove the subsidized price of wheat from down-country). Wheat is also a fodder crop. Abouttwo-thirds of livestock nutrients come fromcereal straw. The yield of fodder from a localvariety of wheat can be more than twice thatfrom an improved variety. Moreover, winterwheat can substantially increase fodder pro-duction if uncontrolled livestock grazing can bemanaged. And increasingly, seasonal laborrequirements will be a factor, as will compet-ing demand for more high-value crops. It istherefore not so obvious in a systems contextwhat would happen to wheat production ifimported wheat were not subsidized. This is anissue warranting some farm managementresearch that incorporates risk.

13. In 1997, the constituents of farm incomeswere as follows: crops/vegetables, 34 percent;fruit, 12 percent; forestry, 14 percent; and live-stock, 40 percent.

14. These include linkages between suchthings as altitude, slope/radiation, sunlight hours,soils (old versus new), water source (includingglacial water temperature), market distance, endof season (crunch) period of harvest/grazinganimal return/early snowfall, boundary wall ani-mal control (winter wheat not being commonlysown because of livestock damage, yet yieldinghuge fodder gains over spring planting), and sea-sonal prices (especially including fodder prices).

15. For example, why is there a very weakrelationship between price and quality in live-stock? Is this why forage is often sold? Is the intro-duction of improved animals changing this?

E n d n o t e s

1 3 5

16. These are areas of dry, rocky rangelandextending up to the high mountain pasturesabove the irrigation channels. The area is ahigh-altitude desert with limited forage growthfrom direct rainfall. But there is a summer flushwatered by glacier and snow melt providingseasonal pasture—the source of most of theannual livestock weight gain.

17. In 1999, the AKRSP embarked on a three-year international collaborative research project,funded by the European Economic Communityand involving seven national and multinationalinstitutions. It was spearheaded by a team fromthe Macaulay Land Use Research Institute(MLURI) in Aberdeen, U.K. This project offerspromise for understanding constraints in a sys-

tematic manner, mapping seasonal availability oflivestock feed, considering not only inter- andintra-cropping zone linkages but also spatialaspects of village locations and seasonal use ofpastures vis-à-vis stall-feeding during the win-ter months.

18. Typically, about 20 varieties enter thefirst year. These are narrowed to about 10 prom-ising selections the next year, and finally toabout 6 varieties that are offered to farmers.

19. For maize, PPB should ultimately be tar-geted at all zones where this crop is grown, asrecommended by a consultant study.

20. Noted also by the JRM.21. Crop damage from silt load in irrigation

water is often a problem.

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

1 3 6

AKRSP (Aga Khan Rural Support Program). 2001.“A Strategy for the Future.” Discussion Paper.AKRSP, Pakistan.

——. 2000a. Toward A Shared Vision, Partner-ships for Development. Annual Report 2000.AKRSP, Pakistan.

——. 2000b. An Enquiry into InstitutionalMaturity of Village and Women’s Organisa-tions. Results of the Institutional Develop-ment Exercise 1998 (A Process MonitoringApproach). Policy and Research Section,AKRSP Core Office, Gilgit, Northern Areas,Pakistan.

——. 2000c. An Assessment of Socio-EconomicTrends and Impact in Northern Pakistan(1991–1997). Findings from AKRSP’s FarmHousehold Income and Expenditure Surveys.Policy and Research Section, AKRSP CoreOffice, Gilgit, Northern Areas, Pakistan.

——. 2000d. A Synthesis of the Findings fromthe Impact Studies on Land DevelopmentProjects. Policy and Research Section, AKRSP,Pakistan.

——. 2000e. A Synthesis of the Findings fromthe Impact Studies on Power Generation Pro-jects. Policy and Research Section, AKRSP,Pakistan.

——. 2000f. A Synthesis of the Findings fromthe Impact Studies on Communication Pro-jects. Policy and Research Section, AKRSP,Pakistan.

——. 2000g. A Comparative Cost Analysis ofAKRSP’s Investment in Infrastructure Projectsin 1998–1999. Policy and Research Section,AKRSP Core Office, Pakistan.

——. 2000h. “Microfinance Program OperationalReview.” MicroBanking Bulletin 5.

——. 2000i. “Performance Indicator Report.”Microfinance Group, Pakistan.

——. 2000j. “Credit Operations Report.” AKRSPMicrofinance Section, Pakistan.

——. 2000k. “Engineering Impact Studies.” Pol-icy and Research Section, Pakistan.

——. 2000l. “An Impact Study of the Water Sup-ply Scheme, WO Narbu-churg (DistrictSkardu).” Baltistan Regional Program Office.

——. 2000m. “A Regional Comparative CostAnalysis of AKRSP’s Investment Infrastruc-ture Projects.” AKRSP, Pakistan.

——. 2000n. Enterprise Development AnnualReport 2000. Core Office, AKRSP, Pakistan.

——. 2000o. Synthesis of the Findings fromImpact Studies. Policy and Research Section,AKRSP, Pakistan.

——. 1999. Joining Hands in Development,Women in Northern Pakistan. Annual Report1999. AKRSP, Pakistan.

——. 1998a. The Challenge of Poverty. AnnualReport 1998. AKRSP, Pakistan.

——. 1998b. “Institutional Maturity Study.”AKRSP, Pakistan.

——. 1997. Review of the Dutch Funded Pro-gramme of AKRSP in Baltistan (January 1986to December 1997). Revised Report. Moni-toring, Evaluation and Research Section, AgaKhan Rural Support Program, Skardu,Baltistan, Northern Areas, Pakistan.

——. 1982 to 1983. First Progress Report Decem-ber 1982 to April 1983. Aga Khan Rural Sup-port Program. Gilgit, Northern Areas, Pakistan.

Ahmad, Nazir, Ingrid Nyborg, and GulcheenAqil. 1998. High Altitude Integrated NaturalResource Management. Report No. 5, Gen-der, Resource Management and LivelihoodSecurity. Aga Khan Rural Support Program,Baltistan, and NLH Agricultural University ofNorway.

Akram, Mohammad Usman. 2001. “Devolutionof Power in Northern Areas: The Success ofV/WOs in Dehi Councils in Gilgit.” AKRSP,Pakistan.

Ali, Mohammad. 1998. Socio-Economic and Cul-tural Changes in Pari, Kharmang, Baltistan.Aga Khan Rural Support Program, Baltistan.

Blackburn, James, Robert Chambers, and JohnGaventa. 2000. “Mainstreaming Participationin Development.” OED Working Paper SeriesNo. 10. Washington, D.C.: OED.

Conway, Gordon R., Tariq Husain, Zahur Alam,and Mian M. Alim. 1987. “Rapid RuralAppraisal for Sustainable Development: Expe-riences from the Northern Areas of Pakistan.”Paper presented at the International Institutefor Environment and Development Conferenceon Sustainable Development, April 28–30.

DAC/OECD (Development Assistance Commit-tee, Organization for Economic Cooperationand Development). 2001. “Glossary of Eval-

1 3 7

REFERENCES

uation and Results-Based ManagementTerms.” Paris.

DFID (Department for International Develop-ment, U.K.). 2000. “Mid-term Review.” London.

——. 1999. “Annual Review 1999.” London.——. 1998. “Report on the Micro-hydro Com-

ponent of AKRSP, Chitral District.” London.Ellingsen, Liv, and Maqsood Khan. 1998. High

Altitude Integrated Natural Resource Man-agement. Report No. 6, Information and Doc-umentation. Aga Khan Rural SupportProgram, Baltistan, and NLH Agricultural Uni-versity of Norway.

Hofmann, Kathrin C., Iqbal Hussain, Age Nyborg,Veronika Seim, Oystein Holand, Per Wegge,Ulrik Motzfeldt, and Mohammad Abbas. 1998.High Altitude Integrated Natural ResourceManagement. Report No. 3, Pasture, Livestockand Diversity. Aga Khan Rural Support Pro-gram, Baltistan.

HTS Consultants, The Overseas DevelopmentGroup, Intermediate Technology Consultants,and Ian Tod and Associates. 1999. Develop-ment of Dried Apricot Marketing in Hunza/Gilgit Region. Aga Khan Rural Support Pro-gram, Gilgit Region, Northern Areas, Pak-istan, and Department of InternationalDevelopment, UK (DFID). Final Report.

Husain, Tariq. 1986. Wheat in the High Moun-tain Valleys of Gilgit. Rural Sciences ResearchProgram, Report No. 2, Aga Khan Rural Sup-port Program, Gilgit, Northern Areas, Pakistan.

Ibrahim, Durr-E-Nayab and Sabiha. 1994. “TheAppropriateness of a Community-based Pro-gramme: A Case-study of the AKRSP in TwoVillages of Gilgit District.” The Pakistan Devel-opment Review.

Joint Monitoring Mission. 1997. “Report.” AKRSP,Pakistan.

Joint Review Mission. 1999. “Final Report.”AKRSP, Pakistan.

Khan, Shaib Sultan. 1989. The Dynamics of RuralDevelopment. Aga Khan Rural Support Pro-gram, Gilgit, Northern Areas, Pakistan.

Khan, Shoaib Sultan, Husain, Tariq. n.d. “Prin-ciples and Implementation for Small FarmerDevelopment.” AKRSP, Pakistan.

Khwaja, Asim Ijaz. 2000. Can Good Projects Suc-ceed in Bad Communities? Collective Action

in the Himalayas. Harvard University Fac-ulty Research Working Paper RWP01-043.Cambridge, MA.

——. 1999. “Revisiting Rural InfrastructuralInvestments: A Means to an End or an Endin Themselves.” MIES Study 1999. Part 1.AKRSP-Baltistan, Pakistan.

Moffat, Linda. 2001. “AKRSP-Baltistan: Organi-zational Gender Assessment.” Aga Khan Foun-dation, Canada.

Monitoring, Evaluation, and Research (MER) Sec-tion. n.d. “Evolution of Community-BasedInstitutions in Giglit Region.” AKRSP, Pakistan.

OED (Operations Evaluation Department). 2002.Social Funds: Assessing Effectiveness. Opera-tions Evaluation Study Series. Washington,D.C.: World Bank.

——. 1996. The Aga Khan Rural Support Pro-gram. A Third Evaluation. A World BankOperations Evaluation Study. Washington,D.C.: World Bank.

——. 1990. The Aga Khan Rural Support Pro-gram in Pakistan: A Second Interim Evalua-tion. A World Bank Operations EvaluationStudy. Washington, D.C.: World Bank.

Oza, Apoorva. 2000. “Scaling-up Community-based Development. Lessons from Aga KhanRural Support Program (India).” AKRSP, India.

Pakistan Government, National ReconstructionBureau, Executive Secretariat. 2000. “LocalGovernment Plan 2000.” Islamabad.

Rehman, Mujeeb Ur. 1997. The Strategy Paper forthe Enterprise Development Division. Enter-prise Development Division, The Aga KhanRural Support Program, Chitral.

Seeley, Janet. 2000. “Consultancy on Support forAKRSP Gender Strategy.” AKRSP, Pakistan.

Stiles Associates, Inc. 2001. “Aga Khan RuralSupport Program, Phase IV (AKRSP IV).” DraftMonitoring Mission Report Feb–Mar 2001.Submitted to Canadian International Devel-opment Agency. Project No: A-020522. Ottawa.

Task Force on Sustainability. 2000. Sustainabil-ity and AKRSP: Plans for the Future. AKRSP,Pakistan.

Tetlay, Khaleel, and Muhammad A. Raza. 1998.From Small Farmer Development to Sustain-able Livelihoods: A Case Study on the Evolu-tion of AKRSP in Northern Pakistan. Aga

T h e N e x t A s c e n t : A n E v a l u a t i o n o f t h e A g a K h a n R u r a l S u p p o r t P r o g r a m , P a k i s t a n

1 3 8

Khan Rural Support Program, Gilgit, North-ern Areas, Pakistan.

Tod, Ian, and Associates. 1999. A ParticipatoryApproach to Cereal Improvement. Consul-tancy Report No. 11, Aga Khan Rural SupportProgram, Gilgit Region, Northern Areas, Pak-istan, Department for International Develop-ment, UK (DFID).

UNDP (United Nations Development Program).1998. “Interim Evaluation of the NationalRural Support Program.” Islamabad, UNDPPakistan.

Walt Whitman. 1900 “Crossing Brooklyn Ferry.”In Leaves of Grass. Philadelphia, PA: DavidMcKay. (Cited in AKRSP 2001.)

Whiteman, P.T.S. 1985. A Technical Report ofAgricultural Studies (1982–1984) in GilgitDistrict, Northern Areas, Pakistan. Aga KhanRural Support Program, Integrated Rural

Development/Department of Agriculture,Gilgit.

Wood, Geof. 1996. Developing the Art and Avoid-ing the Totem: A Strategy for Social Organi-zation in AKRSP. Center for DevelopmentStudies, University of Bath and AKRSP.

World Bank. 2001. Pakistan Country AssistanceStrategy—Progress Report. Washington, D.C.

——. 1999. The Islamic Republic of Pakistan fora Poverty Alleviation Fund Project. Washing-ton, D.C.

The World Conservation Union, United NationsDevelopment Program, and Ministry of Envi-ronment, Local Government and Rural Devel-opment. 1995–1999. Maintaining Biodiversityin Pakistan with Rural Community Develop-ment. Project Process Evaluation, LessonsLearned during the PRIF (Pilot Phase). Gilgit,Pakistan: IUCN.

R e f e r e n c e s

1 3 9

The Operations Evaluation Department (OED), anindependent evaluation unit reporting to the WorldBank’s Executive Directors, rates the developmentimpact and performance of all the Bank’s completedlending operations. Results and recommendations arereported to the Executive Directors and fed back intothe design and implementation of new policies andprojects. In addition to the individual operations andcountry assistance programs, OED evaluates the Bank’spolicies and processes.

Summaries of studies and the full text of the Précisand Lessons & Practices can be read on the Internet athttp://www.worldbank.org/html/oed

How To Order OED PublicationsOperations evaluation studies, World Bank discussionpapers, and all other documents are available from theWorld Bank InfoShop.

Documents listed with a stock number and price codemay be obtained through the World Bank’s mail orderservice or from its InfoShop in downtown Washington,D.C. For information on all other documents, contactthe World Bank InfoShop.

For more information about this study or OED’sother evaluation work, please contact ElizabethCampbell-Pagé or the OED Help Desk.

Operations Evaluation DepartmentPartnerships & Knowledge Programs (OEDPK)E-mail: ecampbellpage@worldbank.orgE-mail: eline@worldbank.orgTelephone: (202) 458-4497Facsimile: (202) 522-3200

Ordering World Bank PublicationsCustomers in the United States and in territories notserved by any of the Bank’s publication distributors maysend publication orders to:

The World BankP.O. Box 960Herndon, VA 20172-0960Fax: (703) 661-1501Telephone: (703) 661-1580The address for the World Bank publication databaseon the Internet is: http://www.worldbank.org (selectpublications/project info).E-mail: pic@worldbank.orgFax number: (202) 522-1500Telephone number: (202) 458-5454

The World Bank InfoShop serves walk-in customersonly. The InfoShop is located at:

701 18th Street, NWWashington, DC 20433, USA

All other customers must place their orders through theirlocal distributors.

Ordering by e-mailIf you have an established account with the World Bank,you may transmit your order by electronic mail on theInternet to: books@worldbank.org. Please include youraccount number, billing and shipping addresses, the titleand order number, quantity, and unit price for eachitem.

OPERATIONS EVALUATION DEPARTMENT PUBLICATIONS