Human Development Report

Français | Español |

Dear HDR Network Members,

Indonesia is just about to initiate a number of Provincial Human Development Reports as part of our programme to support decentralization and strengthen provincial-level governance and strategic development planning.

The Indonesian Provincial HDRs will promote policy dialogue on key development challenges and help create a policy consensus between provincial and district governments around specific themes.

The Provincial HDRs will also monitor the performance of district-level governments through disaggregated HDI and human development indicators. The reports will be combined with other initiatives to monitor the performance of district governments, including an award initiative, possibly some kind of balance scorecard, and other tools.

We are very much interested in examples of how other countries are using sub-national HDRs and/or disaggregated HD data and monitoring systems to strengthen decentralization and local governance, to promote policy dialogue between provincial and local governments, to guide resource allocations and budgeting, and to assess local government performance in terms of development outcomes.

In addition, any information or ideas about consultative process, institutional anchoring at national and provincial level, centralized quality control, and links with national HDRs would also be very helpful.

We understand Brazil and Egypt are good examples, so any detailed information there or from other countries would be highly appreciated.

Many thanks

With best regards

Hakan

Håkan Björkman
Country Director, Indonesia
United Nations Development Programme

Chers membres du réseau RDH,

L’Indonésie est sur le point de lancer un certain nombre de Rapports provinciaux sur le développement humain dans le cadre de notre programme visant à appuyer la décentralisation et à renforcer la gouvernance au niveau des provinces de même que la planification du développement stratégique.

Les RDH provinciaux indonésiens favoriseront le dialogue politique sur les problèmes clés de développement, et contribueront à réaliser un consensus politique entre les gouvernements provinciaux et des districts autour de thèmes particuliers.

Les RDH provinciaux permettront en outre de contrôler les résultats des gouvernements au niveau des districts grâce à des indices de développement humain (IDH) ventilés et des indicateurs de développement humain. Les rapports seront combinés avec d’autres initiatives en vue de contrôler les résultats des gouvernements départementaux, au nombre desquelles figurera la création d’un prix, un éventuel système de fiche récapitulative de suivi et d’autres outils.

Nous sommes très intéressés par des exemples sur la manière dont les autres pays utilisent les RDH sous-nationaux et/ou les données ventilées en matière de DH ainsi que les systèmes de contrôle en vue de renforcer la décentralisation et la gouvernance locale, de promouvoir le dialogue politique entre les gouvernements provinciaux et locaux, en vue d’orienter les allocations de ressources et la budgétisation, et pour évaluer les résultats des gouvernements locaux au niveau du développement.

En outre, toute information ou idée concernant le processus consultatif, l’ancrage institutionnel aux niveaux national et provincial, le contrôle centralisé de la qualité et les liens avec les RDH nationaux seraient particulièrement utiles.

Nous savons que le Brésil et l’Egypte peuvent fournir de bons exemples à cet égard, et toute information détaillée de ces pays et d’autres seraient les bienvenus.

Un grand merci à l’avance.

Meilleures salutations

Hakan


Håkan Björkman
Directeur de pays, Indonésie
Programme des Nations Unies pour le développement

Estimados Miembros de la Red IDH,

Indonesia está a punto de iniciar sus Informes de Desarrollo Humano Provinciales, como parte de nuestro programa de apoyo a la descentralización y fortalecimiento de la gobernabilidad y la planificación estratégica en el ámbito provincial.

Los IDH Provinciales de Indonesia promoverán un diálogo en políticas sobre retos clave y contribuirán a crear un consenso político entre los gobiernos provinciales y distritales alrededor de temas específicos.

Los IDH Provinciales también monitorearan el desempeño de los gobiernos en el ámbito distrital mediante la desagregación de indicadores de desarrollo humano y del IDH. Los Informes estarán combinados con otras iniciativas para monitorear el desempeño de los gobiernos distritales, incluyendo una iniciativa de reconocimientos, posiblemente algún tipo de ficha de evaluacion de resultados, y otras herramientas.

Estamos muy interesados en ejemplos de cómo otros países están utilizando Informes de Desarrollo Humano sub nacionales y/o datos desagregados y sistemas de monitoreo del Desarrollo Humano. para fortalecer la descentralización y la gobernabilidad local, promover diálogo sobre políticas entre los gobiernos provinciales y locales, guiar la asignación de recursos y la elaboración de presupuestos y evaluar el desempeño del gobierno local en términos de resultados en el desarrollo.

Además serían útiles, cualquier información o ideas acerca del proceso consultivo, vinculación institucional en los ámbitos nacional y provincial, control de calidad centralizado y vínculos con los IDH Nacionales.

Entendemos que Brasil y Egipto son buenos ejemplos, de manera que cualquier información detallada de esos países o de otros países, será muy apreciada.

Muchas gracias.

Con mis cordiales saludos,

Hakan

Håkan Bjorkman
Director de País, Indonesia
Programa de las Naciones Unidas para el Desarrollo

Dear Hakan,

The Indonesia initiative with provincial HDRs sounds like we will have a lot to learn from!

In Turkey (another Middle Income Country) we have been generating HD data disaggregated by province and gender since 1996 through our National Human Development Report. Our experience with using disaggregated HD indicators includes policy development and advocacy for reduction of inequalities between regions and provinces in Turkey.

More recently (2005) we have employed HDI in our fiscal decentralization (FD) work where we incorporate human development concerns into inter-governmental revenue sharing and planning of municipal and provincial expenditures, providing a guide or benchmark for resource allocations and budgeting.

In Turkey, inter-governmental revenue sharing, until now, was based solely on population size of the provinces and the municipalities. This system perpetuates a transfer system which does not always support poverty reduction and HD goals. In other words, the fiscal equalization outcome of FD is sub optimal, perpetuating the wide HD disparities in our country.

In order to remedy this situation, we helped design a system for distribution of the resource pool using the following criteria:

EDUCATION: student/teacher ratio and female schooling rate (education of the girl child is a big challenge in Turkey)

SOCIAL: HDI and human poverty rates and some select MDG indicators such as infant mortality

MACRO: GDP/capita and per capita bank deposits (to measure access to financial markets and opportunities; and share of agriculture in GDP

The system we helped design is entitled “Equalization (that is between provinces) Via Fiscal Decentralization” for poverty reduction and elimination of inequalities. This paper describes the system for your ease of reference. The government is currently reforming its inter-governmental revenue sharing norms in a more progressive manner.

You may also find of interest our policy analysis and recommendations for reduction of regional development disparities again employing HD disaggregated by provinces. The executive summary is attached for your ease of reference.

Hope this is useful. We wish you success in your provincial HDR initiative and look forward to hearing more on it.

With kind regards,

Yesim

Yesim M. Oruc Kaya
Senior Program Manager
UNDP - Turkey  

Dear Yesim,

Thank you for your e-mail. It is good to know that you have used the HDI in your fiscal decentralization (FD) work, where you incorporated human development concerns into inter-governmental revenue sharing and planning of municipal and provincial expenditures, providing a guide or benchmark for resource allocations and budgeting. It will be quite useful to others if you indicate as to how you balanced, in terms of resource allocation and expenditures, as for example, among municipalities or among provinces, between needs and performance of those entities. For example, did more resources were allocated or higher expenditures were undertaken in municipalities and provinces with low HDI (need-based) or high HDI (performance-based)?

Kind regards,

Selim

Selim Jahan
Senior Adviser and Cluster Leader
Poverty Group, Bureau for Development Policy
UNDP NY
http://www.undp.org/policy

Dear Håkan,

I am delighted to learn that Indonesia has decided to work on provincial human development reports, with the intent of using these reports to promote policy dialogues, evolve a much needed consensus among provinces (some of which are resource rich, and others poor) on equitable distribution of regional resources, and on the basis of disaggregated human development indicators, at the district level, monitor performance and progress of sub-national governments.

There has been a radical paradigm shift in development thinking - placing people at the centre of development - displacing the dominance of GDP per capita - that has been brought about by UNDP through its global HDRs.

Hats off to you for initiating what a large and diversity country needs - sub-national HDRs, notwithstanding the value and usefulness of the national HDR (Indonesia's national HDRs have been recognised and awarded for quality of analysis).

You mention Brazil and Egypt, quite correctly, for innovations in those countries in producing and getting local governments to act upon the analysis contained in sub-national reports.

I would suggest that the India experience, which is much closer geographically and culturally to Indonesia, and is very similar in its diversity and regional disparities to India, would be an important reference for your purposes. 

A significant number of people, inspired by Mahbub ul Haq, who have contributed to the elaboration of the concept of human development happen to hail from India - Amartya Sen, Sudhir Anand, Meghnad Desai. India has an enormous interest in the concept of human development, and the richness of its intellectual resources related to the preparation of HDRs.

India's Human Development Resource Centre, UNDP India has been kept busy by the sheer size and scale of challenges in a country with over a billion people, and phenomenal diversities.

And India's Solution Exchange has phenomenally capitalised on the intellectual resources, highly disproportionate to its poverty, that India has.

It is interesting that it was India that hit upon the usefulness of producing a provincial HDR for the first time.  A group of young activist/academics persuaded the State Government of Madhya Pradesh to produce a HDR, which was published in 1995. 

UNDP stepped in to assist that team in organizing a workshop in the state capital, Bhopal, to which senior officials from other provinces were invited, with the hope that other provinces would emulate the Madhya Pradesh example. That hope has now been richly realized.

Incidentally, some of key persons associated with that pioneering Madhya Pradesh team also played a key part in the preparation of the 2005 Chattisgarh State HDR, which won the Human Development Award for the most participative HDR produced.  This report is the distillation of village level reports (Jan Rapat,people's report, as they are called) prepared in 19,128 villages of Chhattisgarh. Village committees reviewed them and endorsed them. Subsequently, bottom-up, district reports were prepared, on the basis of a sample (nearly 3,000) of the large number of village reports, selected on the basis of clearly set out criteria. Apart from quantitative data of the usual kind, the uniqueness of this report is the categorisation of people's qualitative perceptions about their own human development. 

As you know, there has been a debate on the HDR network on whether HDRs should be 'independent' (as is the global HDR) or 'owned' by governments. 

One consequence of the State government of Madhya Pradesh taking the initiative to produce a provincial HDR on its own is that all State HDRs in Inida are 'owned' by the governments concerned. UNDP's support and involvement has certainly helped to ensure professional integrity in data gathering and analysis, and editorial independence through the involvement of reputable experts whose opinions command widespread public respect. 

Ownership and official status commanded by these provincial reports has enormously helped in the actualisation to move the HDR forward, from analysis to action. 

The State HDRs of India, I submit, have much more practical utility than all the wisdom and implict policy advice that is contained in the writings of Amartya Sen.  India is proud of him, the Nobel Prize recognizes the value of his work to the world at large. But at the district and village levels, where officials need an understanding and internalisation of human development, to motivate them, and happen to hold the power to act on that understanding, it is the state-owned HDR that matters most.

Desisting from writing what could be an entire dissertation, I would urge you to visit the website of the India HDRC, and follow each one of its links. You will not be disappointed.

It contains answers to many challenges that will be encountered in computed provincial and district-level HDI. 

How to compute one component of the HDI - per capita GDP - at provincial and district levels, is only one of them. How to get gender-disaggregated data?
Or data disaggregated by religion, ethnic group, caste?  It's much more easy to produce a global or national HDR, and much more difficult to produce them at the sub-national level. 

There needs to be an investment in curriculum development, and also enormous capacity development for data collection and analysis.

You will find resources in the UNDP India website that deal with such issues.  It also has a strong focus on gender equality - and that should interest Indonesia a great deal.

As Maxine Olson, UNDP Resident Representative, suggested in 2004, that "while over 400 Human Development Reports have been prepared globally till now ... this is the first time that a systematic follow-up of the messages of the HDRs is being undertaken in the form of a formal project with the apex planning body of the country. This is another "first" to India's credit in its commitment to human development."

South-South Cooperation between India and Indonesia could not be stronger than in the preparation of provincial HDRs, and in the follow-up policy and programme actions.

Warm regards, Sudarshan

R. Sudarshan(Mr)
Policy Advisor-Legal Reform and Justice
UNDP Regional Centre in Bangkok
http://regionalcentrebangkok.undp.or.th

[Facilitator's note: Please find responses from Joachim Nahem (Oslo Governance Centre, UNDP) and Jochen Jesinghaus (European Commission Joint Research Centre)]

Joachim Nahem, Oslo Governance Centre, UNDP, wrote:

Dear Håkan,

We produced a paper with HDRO on “NHDRs and the Use of Democratic Governance Indicators” http://www.undp.org/oslocentre/docs06/NHDR.pdf which surveyed all NHDRs (up until 2004) for instances of governance indicators at the national and sub-national levels.

You will find that several of the case studies have made use of sub-national data, ie disaggregating data by regions, provinces, municipalities etc. as a way of highlighting governance/hd inequalities. We ague that the NHDR framework has tremendous potential to pro­duce governance indicators that are: poverty and gender focused, policy-action oriented, nationally owned and participatory, and methodologically sound. However, NHDRs need to collected governance data more consistently over time thus enabling policy-makers, media and civil society to monitor progress/regress and identify in greater detail priority-attention areas in countries.

The Indonesian initiative is very interesting and we hope that our work on governance indicators can be of use http://www.undp.org/oslocentre/flagship/governance_indicators_project.html.

Joachim

Joachim Nahem
Governance Specialist

Oslo Governance Centre
Democratic Governance Group
Bureau for Development Policy
United Nations Development Programme (UNDP)

Jochen Jesinghaus, European Commission Joint Research Centre, wrote:

Dear Selim, Yesim and Håkan,

We recently published a sub-national indicator set for Tanzania' provinces, see http://esl.jrc.it/dc/tza_dist/index.htm - and we would like to add more examples of good practice, especially some which match closely the MDG set.

The dashboard tool used for this site has been specifically designed for complex indicator sets, and such a site, with an interactive interface including colour-coded maps, can be created in a few minutes; provided that the data sit in a single Excel sheet with provinces in the rows and indicators in the columns.

If you have datasets in Excel, please contact me directly at Jochen.Jesinghaus@jrc.it

Best regards,
Jochen Jesinghaus
European Commission Joint Research Centre
Italy

P.S.: Background on the tool: http://en.wikipedia.org/wiki/Dashboard_of_Sustainability

It is really good to learn that colleagues in Indonesia are looking at sub-national HDRs (provincial and perhpas local/ city level also?) as a point d'appui for interventions on the ground in the areas of decentralisation and federal governance.

I have a few suggestions for Håkan ...

a) It is important to prepare/translate the HDR in the local language/s. It is also important to build up some hype/ visibility about the exercise, so that people do not see it as a routine document. Highest level political buy-in has helped us a lot in India.

b) The state/ district level HDRs gain credibility both with government and civil society, when they use objective, non-departmental data to provide an independent baseline. therefore, it would be important to invest time and money in getting data on outcomes rather than inputs, preferably from large national databases. So data on education, health, income and poverty must conform as closely as possible to the indicators used for the national HDR. However, the subnational HDI, is often based on the "art of the possible", and since index construction in any case involves "information loss", one need not be a purist.

b) Income and poverty data at the district (and below) level is often difficult to get or construct. But it would be important to do so in order to have a district/ local level HDI. For all the criticism thrown at it, the HDI is the "magic trick' that gets you visibility, debate on ranks and value -- all the things that get the intial "hook" for the HDR. Provincial HDRs, if they have a district/ local level HDI, can spark competitive interest of politicians, media and officials, both as a "rank" and and also as a possible basis for resource allocation.

c) "Participatory process" meaning inclusion and involvement of a wide range of stakeholders esp, NGOs, local language media, researchers, is important because it builds the constituency for the HDR exercise from the preparatory phase itself. However, it requires attention to capacity building -- development of training modules, Training of Trainers, local level animators/ facilitators, and above all serious number-crunching and qualitative analysis capacity in the team tasked with preparing the document itself. The motto of "use local capacity to build local capacity" is useful, even if that takes more time.

Often, in the case of local level HDRs, such capacity may not be easily available/ known locally. In which case, it would be important to use networks and reputed national consultants, not as "fly in and out" manner, but a sustained deployment for the duration of the exercise. One can put technology and content development in local idiom to good use. For instance, in the south Indian state of Karnataka, the VSAT network of the State Training Institute, is being used to do an "interactive classroom" in the local language, for nearly 17000 village/ward council members in 4 low HDI districts for which district HDRs are being prepared.

d) it helps to have a practical thrust such as financing of human development, social justice, women's security, development of religious/ ethnic minorities  or any such direct concern of planners and implementors. The provincial HDR exercise should have a direct (and visible) linkage with a pragmatic outcome of state policy/ public sphere.

Gurus of human development such as Amartya Sen have emphasised the "agency" aspect of the HDR exercise, as vehicle for changing mindsets. It would be useful to buttress the HDR preparation with films, media articles, public lectures, legislators' forums, etc. This would create "demand pull" so that the HDR is rooted in fertile soil of public interest.

These are some lessons and issues we have grappled with in our work. In case colleagues in Indonesia (and elsewhere) are interested in India's follow up to State HDRs, they could get in touch with us. A variety of resources, including ToT manuals, training modules, films, and reports are available for those interested in using subnational HDRs to help improve the situation. In India the operationalisation of HD through subnational HDRs is very much "work in progress" and we need to learn more from other country experiences.

There is a lot of interest in Indian states (where HDR messages are under implementation) regarding the excellent work under way in Brazil and Egypt as also other countries. It would be good if colleagues on the surf-hdr net share experiences on follow up and policy impact, esp. with regard to capacity development work and advocacy.

I wish Håkan and colleagues the very best.

Regards

Suraj

I am also enclosing for perusal, a brief note on some follow up activities in India. Comments and feedback from network members on this, also on what more could be done, would be most helpful.

Translation I Original Message

Translation

The NHDR Team of Costa Rica has been working closely with the University of Costa Rica to develop a family of HD indicators.

Our first experience came up with the NHDR 2005 on citizen security and HD in Costa Rica, where we incorporated a sub regional calculation of the HDI into the national HDI. We also built a “security indicator” at the sub regional level (Indice de seguridad cantonal, ISC) and its impact on HD (Indice de desarrollo humano corregido por seguridad, IDHCS). We have been very successful in the use of those indicators, as they are being used for several projects and analysis in the country.

During last year, the focus of our team has been the development of the Gender Development index, the Gender Mainstreaming index and the Human Poverty index, at the sub regional level as well. The result of that work will be available in the “Atlas de Desarrollo Humano cantonal” (Sub regional Human Development Atlas) that will be accessible on line next month.

The development of these indicators has not been easy: although Costa Rica has a lot of disaggregated data at the sub regional level available, there are some important gaps and out of date data. Other challenges are also the high mobility of the population between the province where they live and the one where they work or use services (given the small extension of the territory), which makes the calculation even harder. We explain in the NHDR 2005 how we overcame these challenges.

I hope this helps. Please don’t hesitate to contact me if you need further information. Cheers!

Original Message

Estimados colegas.

El trabajo de desarrollar la familia de indicadores del desarrollo humano, y otros, ha sido una tarea que el equipo del INDH de Costa Rica en conjunto con la Universidad de Costa Rica, trabajan para diseñar la metodología idónea de monitoreo de estos indicadores.

Nuestra primera experiencia se brindó con el INDH 2005 titulado "Venciendo el temor, Inseguridad ciudadana y desarrollo humano en Costa Rica", en el que incorporamos el cálculo del IDH cantonal para el país, además de construir un indicador de seguridad (Índice de Seguridad cantonal -ISC-) y el impacto de este último sobre el desarrollo humano ( Índice de Desarrollo Humano corregido por Seguridad -IDHCS-). La información disponible para estos indicadores a resultado con muchos éxitos, ya que a nivel institucional como de políticas públicas se toman en cuenta para el desarrollo de proyectos y análisis del país.

Durante el último año el trabajo del equipo se destino a desarrollar además del IDH , el Índice de Desarrollo relativo al Género (IDG), el Índice de Potenciación de Género (IPG) y el Índice de Pobreza Humana (IPH), también a nivel cantonal. El resultado de este trabajo estará disponible en la publicación del "Atlas del Desarrollo Humano Cantonal" que publicaremos el próximo mes y estaremos habilitado en línea.

La tarea de construir estos indicadores no ha sido fácil; si bien Costa Rica cuenta con un amplia gama de información desagregada a nivel cantonal, las necesidades en algunos campos aún se mantienen, ya sea por ausencia o por falta de periodicidad. Sumando a ello, al ser un país con una extensión territorial pequeña, la movilidad de la población entre el cantón de residencia y sus vecinos para desarrollar actividades y suplir servicios, principalmente en el centro del país, se convierte en una variable de mucha importancia para considerar, a fin de mejorar las estimaciones. En la publicación del INDH 2005 pueden encontrar la descripción sobre como fueron solventadas.

Espero que pueda ser de utilidad y quedo a la orden para cualquier consulta. Éxitos !!!

Saludos,

Gerald.

Gerald Mora Muñoz
Informe Nacional de Desarrollo Humano
PNUD Costa Rica

[Facilitator's note: Please find responses below from Shaila Khan, UNDP Bangladesh and Seheir Kansouh-Habib, President, AFCS Egypt]

Shaila Khan, UNDP Bangladesh, wrote:

Dear Hakan,

I would like to join Sudarshan in congratulating you for taking up the challenging and the mammoth task. Indeed inspiring to know about the provincial HDR which is underway. While it is a challenging task, it is also an opportunity to localising the MDGs too. I believe the provincial level MDG report could also add value, and have provincial based MDG based planning something you may think through at a later stage. The initiative tells me that Indonesia must have a strong data set both at the central and provincial level. In case of Bangladesh it is difficult to have data at that level –in our case province is almost equivalent to Division -, which is the second tier of the administrative level. I would be interested to know more on the time frame in getting the data set in other words what is the source data that will be used in the preparation of the provincial HDR? Do you have disaggregated data too?

It will be also interesting to learn the experiences/challenges that you may have gone through by now or envisaged.

I would conclude by wishing you all the best for taking such a path breaking initiative. We would certainly like to build on your experience for taking similar initiatives in Bangladesh.

Best of luck and with regards,

Shaila Khan
Policy Support and Advocacy,
UNDP Bangladesh

Seheir Kansouh-Habib, President, AFCS Egypt, wrote:

Dear Colleagues,

Further to the information on the India experience, I also have the privilege to advise that Egypt has been a pioneer in using provincial HDIs back in 1994. It is true that the size and magnitude of Egypt (26 Governorates + Luxor city) can in no way compared with the geopolitical dimension of Indonesia - hence more similarities with India, still the process we had adopted at this time was considered worth being presented to the Compenhagen World Social Summit of 1995. The process management was further described in an article which was published in a peer reviewed academic journal in the USA (I may not have it on the hard disc but it could be faxed if of interest).

What is most important is not only the fact that this exercise caused a redistribution of investments to target previouslay underserved areas, but it was sustained, upgraded and as from 2003, governorates' reports are beeing issued down to the district levels and substantive funds were earmarked to support the poorest amongst them in a large project. Egypt's HDR and sub-national HDRs in Egypt have been flagships of UNDP in Egypt and I think globally.

I am currently a retired UNDP official but am happy to introduce you (in case the link is not yet established) with current officials who are in charge of the ongoing initiatives. Their names are: Dr. Hamed Mobarek - Mrs. Ghada Waly - Mrs. Nahla Zeitoun, as well as the current Principal Investigator for the national HDR since 2004 is Dr. Heba Handoussa

I hope you will find much value in connecting with them and benefitting fromt their experience.

Best,

Seheir Kansouh-Habib
President, AFCS Egypt

Dear Håkan,

I am delighted that Indonesia is initiating provincial HDRs. Sudarshan and Suraj have already provided you with much of the information regarding the Indian initiative. In larger countries with regional variations, provincial HDRs are useful in ensuring greater ownership of the human development approach among policy makers as well as the general public.

In India, encouraged by the experience of their  first provincial HDRs, many provincial governments in India are embarking on their second and third HDRs ( for example, Madhya Pradesh which has 3 state HDRs in 1995, 1998 and 2002, Karnataka, 1999 and 2006, and West Bengal in 2004 currently preparing the second HDR). Most of them are preparing these HDRs on their own without financial support from UNDP as they find these documents useful for policy and planning.

This being the case, and in view of the need to develop capacity at the local level on human development issues and approach, it would be good to think in terms of introducing human development in the course curriculum of schools and universities. We have initiated such an effort in India at the university level, details of which can be found at  http://www.undp.org.in/index.php?option=com_content&task=view&id=105&Itemid=191.

The range could include an introductory lesson at the school level (in India the HDI is explained in Xth Standard books in a simple way), specific optional courses for graduate and under graduate students in universities in the social sciences stream, short-term courses for practitioners such as government officers, NGOs, media and parliamentarians whose understanding and involvement would be crucial in ensuring that the message of human development is widely disseminated.

The results of such an exercise would be a trained local cadre of human development researchers and practitioners in about 3-5 years which would lead to greater sustainability of the valuable efforts you are embarking on.

best regards

Seeta

Seeta Prabhu
UNDP India

[Facilitator's note: Please find responses on Brazil's experience from Eduardo Martins (senior consultant), UNDP Brazil and Marielza Oliveira, RBLAC, UNDP NY]

Eduardo Martins (senior consultant), UNDP Brazil, wrote:

Dear Hakan,

Congratulations on such an important initiative. We have enjoyed great success with a series of sub-national HD reports here in Brazil; I will try to provide a brief account of our experience.

Brazil is a country with continental dimensions - 8.5 million square kilometers, 190 million inhabitants, 27 provinces, and over 5,500 municipalities. In 1996, UNDP Brazil teamed up with IPEA, a government research institute, to produce the first Brazilian HD Report. This report presented the HDI and a host of accompanying indicators for the 27 provinces. The report was well received, and introduced the HD concept to the Brazilian public, but our big breakthrough would come two years later, when UNDP Brazil and IPEA, joined by another government research institution (João Pinheiro Foundation) released the first "Atlas of Human Development in Brazil". The Atlas contained data (about 200 indicators) for the 27 provinces and for all 5,500+ municipalities, and included a custom designed software to provide easy access to the enormous amount of data assembled, allowing users to easily create queries, rankings, maps, reports, charts, etc. Some samples are provided below:

Rankings:

Maps:

The combination of the HDI presentation at the most disaggregated administrative level (municipality) and of a freely distributable tool allowing users to manipulate the data were the key ingredients to the widespread acceptance of the Atlas, and its enormous impact on the formulation of federal and province-level government policies. By making the full database, along with the tools to manipulate it, easily available, we encouraged independent analyses by all sorts of different stakeholders, such as NGOs, state and local governments, research institutions, grass roots organizations, etc. "Experiences with the Development and Use of Poverty Maps - Case Study Note for Brazil", prepared by the Poverty Mapping project (a joint initiative by FAO, UNEP and CGIAR), provides a detailed case study of the impact of the "Atlas" (the full report, with summary case studies for 14 countries, can be downloaded here). Both the Brazilian case study note and the full report are well worth reading and provide a wealth of useful information. For those that do not want to read the full reports, the main findings are summarized below.

Uses, Brazil Atlas

• used in the so-called Alvorada Program for various poverty reduction schemes (e.g., in education, water, health, and infrastructure), affecting the allocation of approximately US$4.2 billion over three years (2000-2002)
• used as a guideline to allocate social development expenditure in the 2001 national budget, affecting the allocation of approximately US$18.4 billion that year
• used by a large insurance agency, Atlântica Boavista, to influence life insurance premiums
• used by General Motors to locate a site for a new auto manufacturing plant
• used by the Federation of Industry and Commerce in the State of Rio Grande do Sul (ranked highest according to the Human Development Index (HDI)) to attract investment
• used by the largest industrial gas company in Brazil, White Martins, to provide youngsters (aged 15-17) with grants to act as community health or human rights agents (US$1.17 million)
• used by the Ministry of Communications to identify poor municipalities for free access to e-government services and online information
• used to allocate free medicines under the state’s “essential pharmacy” program
• referred to in political speeches and by civil society

Impacts

• helped facilitate more transparent and accountable resource allocation and decisions
• resulted in the use of geographic targeting in programs
• ignited media and local policy debate (e.g., reference to the HDI has become common)
• generated demand for higher-resolution and more timely data
• raised awareness of the limits of poverty data (e.g., that may be based on older data)

Driving forces

• demand-driven: to provide high-resolution data (e.g., to respond to decentralization)

Dissemination

• CD version of the atlas includes:
  - user-friendly software to develop user-defined maps, desired indices, and thematic indicators
  - has easy-to-use import and export functions
  - software developed in-country from scratch to avoid expensive royalty fees
  - software runs on commonly available IBM 486 computers and requires little hard drive memory
• CD disseminated at no charge to all libraries and various federal and state agencies, think tanks, and research institutions
• CD available as a free download at UNDP Brazil, IPEA and FJP web sites
• simple, easy to understand, and easy to replicate indicators were intentionally developed
• widespread publicity through newspaper and television coverage:
  - two hours of prime-time television as well as front page newspaper coverage (estimated to have reached 16 million viewers)
  - more than 50 newspapers from every region published editorials

Capacity Building

• high level of in-country capacity: various institutions have been involved in developing numerous versions of the atlas
• widespread knowledge of the HDI
• HDI incorporated in secondary school curricula and university entrance exams

In 2003 a new edition of the Atlas was prepared, and we started producing Metropolitan-region Atlases, with neighborhood level data for the major metropolitan regions in Brazil (the largest metropolitan regions were subdivided in as many as 400 sectors). These were another major step, and show the huge inequalities within Brazil's largest cities - which often are even larger than the inequalities found among provinces or municipalities. One interesting feature of the metropolitan Atlases is the inclusion of aerial photographs and street maps. The photographs and maps (see examples below) created an immediate bond with local populations during meetings to discuss priorities, policies, etc - people could actually see where they lived, and this strenghtened debate participation.

Aerial Image:

Street Map:

All Atlases rely on the decennial censuses for their raw data. Brazil has a strong and independent statistical agency, responsible for the censuses, and we used the full database of individual census returns (with safeguards to protect the identity of respondents, as mandated by Brazilian law) to tabulate the data and calculate the HDI and all other indicators. This provided us with a reliable, homogeneous source of data for all municipalities and neighborhoods, but has the undesirable side effect of limiting data availability to census years. So, data is presented for 1991 and 2000, the latest census years. The ten year intervals prevent the data from being used for monitoring and evaluation, but we do not see this as a major shortcoming - it is widely acknowledged that the HDI is not well suited for this purpose, since it is comprised mainly of stock variables, which change slowly over time. We also produce a yearly series for the 27 provinces only (no municipal data), based on national household surveys.

The latest version of the Atlas of Human Development in Brazil can be downloaded here. It is available only in portuguese, but anyone with a working knowledge of spanish shouldn't have any problems. The Atlases for the metropolitan regions can also be downloaded from UNDP Brazil's web site. Currently availabe are the Atlas of Human Development in Recife, Belo Horizonte, Salvador and Manaus.

Based on our experience, our main suggestions (some of which have already been made by other colleagues here, but are worth repeating) would be:

• use a reliable, consistent database as your main source. This may limit data availabilty to a few years, but if you use data collected by different agencies (especially if you rely on data collected by local governments, rather than a central statistical agency) you may end up with results that are not comparable, and the final results may be questioned;
• for the district-level indicators, get local institutions involved at all stages, beginning at the planning stage. This strenghtens local ownership, and promotes local capacity building;
• make the full database (including raw data used in the calculation of the indices) readily available, preferrably in electronic format. This promotes independent analyses, and allows any interested person or agency (for instance, an opposition political party or research institution) to independently replicate the calculations of the indices, thus validating the results;
• provide data at the most detailed spatial disaggregation you possibly can. People identify more easily with the district or neighborhood levels than they do with more abstract administrative levels. This also creates a "healthy" competition among localities (one interesting side note: we tried to provide a few international comparisons, but quickly found out that people - and governments - were more interested in comparing themselves to neighboring localities than to countries in Europe or somewhere else);
• stay as close to the "standard" HDI methodology as possible. Some adaptations will be unavoidable, but resist the temptation of adding a host of additional dimensions; one of the HDI's main strengths is its simplicity and transparency. Dimensions not contemplated by the HDI may be covered by the inclusion of additional indicators and indices, rather than modifications to the HDI itself;
• beware of using the HDI for monitoring and evaluation, as it is comprised mostly of stock variables and changes slowly over time.

My apologies for the long winded reply, and good luck with this important and exciting initiative; please feel free to contact us if you need any additional information on our experience.

Warm regards,

Eduardo Martins
Senior consultant, UNDP Brazil

Marielza Oliveira, RBLAC, UNDP NY, wrote:

Dear Hakan,

Indeed, what a wonderful initiative, I’m sure the Indonesian provincial reports will contribute immensely to the country’s policy dialogue and development planning processes.

As a former member of the UNDP Brazil team working on the Atlas of Human Development in Recife, the first one in the “city series”, I just wanted to add two things to the explanations and suggestions provided by Eduardo:

1) Although no paper report was published, extensive in-depth analysis of the human development issues facing the area was included in the software, in the form of texts prepared using the database provided via the software. There were mainly 3 types of texts, written to attend the needs of different types of audiences:
i. presentations of the initiative by the RR and the mayor of Recife;
ii. Analytical texts, ranging from an explanation of Human Development and MDGs (as a capacity building materials for those unfamiliar with these concepts) in simple, non-technical language, to sophisticated analyses commissioned by UNDP and by the city of Recife from local think-tanks and academia representatives, examining different SHD issues and proposing policies to overcome problems detected; and
iii. Methodological texts, including metadata descriptions, explanations on how to calculate the HDI and its sub-indices, and information related to the geo-referencing approach adopted (city subdivisions, cartographic data production). Thus, in addition to the database itself, the software offered analyses as well as capacity building tools, to give it the widest possible range of uses.

2) The database was especially constructed to allow not only an understanding of disaggregated Human Development information, but also of the Millennium Development Goals. The dataset includes socio-economic variables related to the MDGs, and the software allows users to view all variables related to a particular MDG (example below: MDG 1, including % of income received by the poorest quintile of the population, average per capita income of the poorest quintile, % people living in extreme poverty, etc). Two data points were included: 1991 and 2000, dates which are significant for the MDGs (the 1991 data points pertain to data collected in 1990, but only made available in 1991). Thus, users also had access to data that gave them the ability, if they so desired, to compute projections of the city’s performance in 2015. We intend to include such projections in future versions.

Inclusion of these 2 features (texts and MDGs data) contributed to the success of the approach, as both provided concrete policy directions for stakeholders to discuss and advocate for, that, while being grounded in a solid and robust statistical foundation, was simple to understand. For example, the mapping tool facilitated demonstration of where in the city investments were most needed in each theme, thus helping to improve the city’s planning process.

I hope this helps. If you need any additional information, please just let us know and we’ll be glad to help. And good luck on your Provincial HDRs, I’m sure they will be a great success!

Warm regards,

Marielza Oliveira
Programme Advisor
Regional Bureau for Latin America and the Caribbean
United Nations Development Program
Millennium Development Goals: Yes we can!