This chapter reports on a particularly Tanzanian research project into the consequences of economic growth for rural poverty. The project explores the poverty dynamics apparent in changing asset portfolios of rural families in Tanzania from the late 1980s onwards (and mainly the 1990s and 2000s). At the time of writing we are half way through data collection for this project. This chapter focuses on the methodological challenges that we have found.
The project is particularly Tanzanian in three important ways. First, it addresses a specifically Tanzanian aspect of debates about poverty dynamics and economic growth and the data deficiencies surrounding both. Second, it is a risky project but is most likely to succeed in a place like Tanzania. The risks arise from its deceptively simple method. We revisit families who were surveyed by researchers in the 1990s, we reinterview them paying particular attention to changing asset bundles, and we explore the reasons behind any changes in prosperity and poverty that we find. However, locating the surveys requires the strong research networks that Tanzania affords, as described in this chapter. The final reason why this is a particularly Tanzanian project is that undertaking this work entails getting to grips with problems of asset ownership and family dynamics that are the bread and butter of rural research in Tanzania.
Our argument is that assets matter when exploring the long-term dynamics of poverty, prosperity and livelihood. By assets we mean, inter alia, land, livestock, houses, household goods and farming implements. Some recent analyses also include education as an asset, blurring the boundaries between social capital and assets (Young 2012). Assets matter because local definitions of what it means to be rich or poor hinge on the bundles of assets to which families have access, and their effective use (see the descriptions in Table 5.1). They capture aspects of poverty and wealth that matter to rural communities. Many measures of poverty, such as poverty lines, which are derived from measures of weekly expenditure, and debates about poverty dynamics around those lines, overlook assets. They may miss important aspects in rural life for that reason. We also argue, however, that measuring change in assets is hard, and we explore the reasons for that in this chapter.