Towards a research framework for poverty monitoring in Tanzania

1.    Introduction
Institutions committed to poverty alleviation must have ideas about why it occurs, why it persists and how it can be overcome to guide their work. Indeed, they have always operated on the basis of specific theories about poverty that reflect their understanding of cultural, social and economic realities. 
Since the second half of the 1980s, public institutions have developed increasingly sophisticated multi-topic surveys as their preferred means to measure, analyse and learn about poverty.  In contrast with single-topic surveys (such as Employment, Income and Expenditure Surveys), these multi-topic Household Surveys are designed to generate information on a wide range of issues intimately linked to household welfare.  At the same time, private development aid institutions and, to a lesser extent, academic institutions were rapidly pioneering a “participatory approach” to developing information and understanding about poverty. 
In their current forms, both methodologies involve poor people in the production of data.  The primary difference between participatory and survey-based research is that the former systematically involves poor people in the analysis of its findings.  It is this analysis, as much as the raw data, which is then synthesised to inform pro-poor policies. 
Some of the advantages to Participatory Policy Research are obvious.  First, data analysis does not depend on speculation by urban elites about the conditions poor people face.  Instead, it is the result of poor people – the “everyday experts on poverty” – reflecting on, theorising about, debating and explaining the world in which they live.  Second, Participatory Policy Research contributes to social democratisation by engaging poor people in policymaking processes.      
On the basis of these characteristics, the Government of Tanzania has decided to make Participatory Policy Research, in the form of Participatory Poverty Assessments (PPAs), a routine part of its Poverty Monitoring System.  As such, the Tanzania PPA Process is firmly enmeshed in national level planning processes.  
The PPA Process is playing a vital role in the increasingly concerted efforts of Government, donors, Civil Society Organisations, the private sector and ordinary individuals to end mass poverty in Tanzania. 
Of course, Government has been concerned with poverty alleviation since Independence.  However, in recent years, plans and procedures to eradicate mass poverty have multiplied in the form of Vision 2025, the National Poverty Eradication Strategy (NPES), the Tanzania Assistance Strategy (TAS), the Medium Term Expenditure Frameworks (MTEFs) and Public Expenditure Review (PER). 
Vision 2025 describes the general level of development the country wants to achieve during the next few decades.  In contrast, the NPES sets more specific poverty reduction targets.  The TAS is a means to coordinate the efforts of GoT with those of the international community in order to reach these goals.  Meanwhile, the Medium Term Expenditure Framework and Public Expenditure Review are important processes enabling Government to prioritise and track the impact of pro-poor public expenditures.
In 2000, Government produced a Poverty Reduction Strategy Paper (PRSP).  It was subsequently approved by Parliament and endorsed by the Executive Boards of the World Bank and IMF.  The PRSP is narrower than Vision 2025, the NPES and TAS in the sense that it covers a shorter time-span and entails more focused objectives.  In other words, Tanzania's new Poverty Reduction Strategy (PRS) is intended to be the key mechanism for coordinating practical initiatives to end mass poverty. 
Government’s Poverty Reduction Strategy and other development efforts depend on knowing whether or not the activities they set in motion are, in fact, improving people’s welfare, how and why.  Therefore, the Government of Tanzania recently established a Poverty Monitoring System (PMS) to provide an institutional framework for rigorous monitoring and evaluation of anti-poverty programmes.  The PMS will: 
  • Guide the timely collection, analysis and dissemination of information 
  • Enable policymakers to assess national poverty alleviation strategies, identify shortcomings and make adjustments as necessary 
  • Allow Government and its development partners to identify particularly successful initiatives so that they can be given adequate support and (if feasible) replicated  
A wide range of stakeholders have been involved in designing the PMS and ensuring that it seeks to provide a comprehensive understanding of poverty trends and their reasons.  In October 2000, a Consultative Workshop was held at the White Sands Hotel in order to accelerate creation of this ambitious System.  Amongst other important conclusions, participants decided that regular Participatory Poverty Assessment – providing “The poor’s perception of trends in poverty and impact of policy changes under the PRS” – should be conducted.  
In 1994/5, the World Bank conducted a PPA in Tanzania.  It illuminated aspects of poverty and wellbeing important to poor people themselves.  It also showed how surveys can distort our understanding of poverty by papering-over the unequal access to economic and non-economic resources experienced by individuals in the same household.  Indeed, findings from this PPA contributed to growing recognition of poor communities and households as heterogeneous units whose members face an array of circumstances demanding a range of policy responses.     
The 1997 Shinyanga PPA worked in a single region.  It built the capacity of local government staff to engage in participatory planning and provided key information for a Human Development Report.
Both these “First Generation PPAs” were designed to collect information about the nature, causes and consequences of poverty from the perspectives of poor people.  They did this well and have provided policymakers with essential information about the complexity, seasonality, etc. of poverty in Tanzania.  Unfortunately neither PPA was designed as a comprehensive process to inform and influence national policy. As a result their impact was limited.
There is still much to learn from PPAs.  Indeed, many basic questions remain unanswered.  Moreover, we need to understand the changing causes and consequences of poverty – research goals the PPA methodology is particularly well suited to pursue.  Thus, the 2000 PRSP stated that:
“The integration of a regular PPA in the PRSP monitoring system... will provide invaluable qualitative data, which will serve to cross-check quantitative data, help us judge the effectiveness of policy measures and more generally help us understand the causal links between the action programmes of the PRSP and changes in poverty.  But most importantly, it will help us listen to the concerns, perceptions and opinions of the poor themselves.”
and the Vice President’s Office explained that PPAs should allow:
“the poor themselves to express their views on how poverty is evolving, what the causes are behind changes in the level and nature of poverty and how different policies and strategies are having an impact on the poor. The data and information coming out of the PPAs will be invaluable to put the quantitative data in context and to enhance our understanding of them”
In sum, the Government of Tanzania hopes that PPAs will improve policymakers’ understanding of poverty and the outcome of poverty alleviation activities as well as open-up political processes by involving ordinary people in the decisions that affect their lives. 
In the context of a comprehensive, sustained process to inform and influence national policy, this new generation of PPAs will do all this and more. 
6.    What does the Tanzania PPA Process look like?                       
The National Poverty Monitoring System is led by an inclusive Steering Committee composed of representatives from government, the private sector, civil society, and the academic/research community.  Technical Working Groups have been established and given responsibility for directing activities under the PMS.  Thus: 
  • The Working Group on Surveys and the Census (led by the National Bureau of Statistics) is establishing a multi-year household survey programme that will provide data on an annual basis for poverty monitoring   
  • The Working Group on Routine Data Collection (led by the President’s Office - Regional Administration and Local Government) is improving the routine data collection systems of local authorities and sector ministries, as well as collating their information outputs 
  • The Working Group on Research and Analysis (led by the President’s Office Planning and Privatisation) is co-ordinating research and analysis of poverty issues to inform the PRSP.  This entails working with an array of stakeholders to set research priorities and ensure high quality investigations.   This Group is also responsible for overseeing implementation of the Participatory Poverty Assessment process   
  • The Working Group on Dissemination and Sensitisation (led by the Vice President’s Office) is developing and spearheading a strategy for the effective dissemination of findings from the Poverty Monitoring System 
The Research and Analysis Working Group is responsible to the PRS Technical Committee for the successful implementation of PPAs.  In order to meet its obligations, the Working Group has formed a PPA Steering Committee to oversee and guide the PPA Process as executed by the Macro-Economy Division of the President’s Office, Planning and Privatisation.   
Participants in the 7th March PPA Stakeholders’ Workshop recommended that, while the process should be executed by a Government Agency, it should be implemented by a Consortium including:
  • Non-profit academic & research institutions
  • National non-governmental organisations
  • International non-governmental organisations
These same stakeholders recommended that the Lead Implementing Partner be a non-profit research or academic institution.  On the basis of this recommendation, the PPA Steering Committee initiated a rigorous, transparent selection process.  Together with several external evaluators, it ultimately chose the Economic and Social Research Foundation (ESRF) to be the Lead Implementing Partner for the 2002/3 PPA.  
Immediately afterwards, the Steering Committee met with ESRF to review applications from other institutions interested in joining the Implementing Consortium.  Their deliberations culminated in the selection of fourteen Implementing Partners (IPs), namely:  
  1. The President’s Office, Planning and Privatisation (PO-PP)
  2. The Ministry of Finance (MoF)
  3. The National Bureau of Statistics (NBS)
  4. Christian Social Services Commission (CSSC)
  5. Concern for Development Initiatives in Africa (forDIA)
  6. The Pastoralists and Indigenous NGOs Forum (PINGOs Forum)
  7. The Institute of Development Studies (IDS), University of Dar es Salaam
  8. Women’s Research and Documentation Project (WRDP)
  9. ActionAid, Tanzania
6.2    Objectives
The first Stakeholders’ Workshop for the PPA Process was held 7th March 2001 in the Courtyard Hotel, Dar es Salaam.  Representatives from Government, donor institutions and civil society organisations attended, discussed and debated the shape to be taken by the PPA Process in Tanzania.  Their conclusions, in combination with Government’s prior expectations, led to the formation of specific goals.  These are: 
  • Enhancing, through in-depth description and analysis, research participants’ and policymakers’ understanding of key poverty issues 
  • Exploring the (a.) different and sometimes competing priority needs of poor people, (b.) likely impact of policies and (c.) tradeoffs and potential compromises between diverse interests in order to develop ‘best bet’ recommendations for poverty alleviation 
  • Facilitating the constructive engagement of civil society in pro-poor policymaking processes
These goals will be refined and supplemented throughout the PPA Process.
6.3    Subject Matter
Each PPA Cycle will focus on a particular subject matter, or “Research Theme,” strategically selected to contribute timely information to key policy debates.  
In June 2001, the R&AWG commissioned a study to identify national-level priority research needs.  This study, entitled Towards a research framework for poverty monitoring in Tanzania, consulted stakeholders and assessed key poverty oriented policy documents.  It concluded that there is especially great need for research on “vulnerability” due, amongst other reasons, to its immense impact on people’s well-being and capacity to rapidly erode improvements made by the PRSP.
On the basis of the methodology’s unique strengths, the R&AWG decided that the 2002/3 PPA should concentrate on this very important Theme.  Priority Research Topics (i.e. broad issues) and Items (i.e. specific subjects) were selected through an inclusive Stakeholders' Workshop conducted on 4th February.  
6.4    Research Sites
On Tuesday, 5th February, the Implementing Consortium for the 2002/3 Participatory Poverty Assessment (PPA) convened a full-day workshop with multiple stakeholders to select “site-areas” (i.e. districts where communities with particular characteristics can be found).  In coming months, Research Teams will visit these districts and – together with Local Authorities – identify appropriate communities for fieldwork.
In light of the Tanzania PPA’s policy role, site-areas were not chosen to illustrate “worst-case scenarios.”  Instead, they were selected to be broadly representative of the diverse circumstances, opportunities and challenges faced by ordinary Tanzanians.  Of course, it is immensely difficult to try capturing something of this diversity in a relatively small number of sites for intensive study.  Indeed, hard choices have had to be made. 
In order to make the best possible choices, Workshop participants formed teams with special expertise around the basic livelihoods supporting communities in Tanzania.  Each team then identified and prioritised the most significant variables of diversity vis-à-vis vulnerability and their assigned livelihood category.  Thus, for example, the variable “Reliable Rainfall” was prioritised by the team responsible for Farming-based Livelihoods but not by the team looking at Urban-based Livelihoods. 
These activities led to the creation of distinct “diversity trees” for:
  • Urban-based Livelihoods
  • Fishing-based Livelihoods
  • Livestock-keeping Livelihoods
  • Farming-based Livelihoods
In some cases, it was relatively easy to then identify sites featuring all the key elements of diversity along a given branch/priority pathway.  In other cases, identifying the most appropriate site has been much more difficult.  The PPA Management Team, therefore, met with additional specialists to confirm the selection and sequencing of some variables and seek further advice on site areas.  This process culminated in the following list of thirty site-areas: 
  •  Bagomoyo District
  • Chunya District
  • Dodoma Rural
  • Handeni District
  • Igunga District
  • Ilala District
  • Iringa Urban
  • Kibondo District
  • Kigoma Rural
  • Kilosa District
  • Kinondoni District
  • Kyela District
  • Lindi Rural
  • Mafia District
  • Makete District
  • Manyoni District
  • Mbulu District
  • Mbulu District
  • Meatu District
  • Muleba District
  • Mwanza District
  • Newala District
  • Njombe District
  • Nkasi District
  • Rufiji District
  • Same District
  • Simanjiro District
  • Songea Rural
  • Tanga Urban
  • Tarime District
(Click here to see where these sites - marked in green - are located in Tanzania.)  
6.5    Timing
PPAs will be implemented in two-year long “cycles” calculated to feed into the PRSP and other policy review processes.  The first of these cycles began in January 2002 with an intensive Training Programme in Participatory Policy Research, conducted by the Institute of Development Studies, University of Dar es Salaam and Development Research and Training (DRT), a Ugandan NGO.  Fieldwork began on 4th March, 2002 and will continue through mid-July.
From July through December 2002, the PPA will undertake further analysis and write-up its research results.  This period will lead to the production of a National Report and an as of yet undetermined number of focused policy briefing papers. 
The entirety of 2003 will be dedicated to encouraging and facilitating the practical use of research results by policymakers and to preparing for the next cycle of research in 2004.