Botswana is among the fastest growing economies in the world in the last two decades (The World Bank 2006). The overall incidence of poverty fell from 60 per cent in 1985/86 to 30 per cent in 2002/03 (Central Statistics Office Botswana 2004). Rural poverty declined from 55 per cent in 1985/86 to 40 per cent in 1992/93. However, the incidence of rural poverty rose to 45 per cent in 2002/03 (Central Statistics Office Botswana 2004). An increased record of rural poverty compared to declining records in non rural poverty during the last decade prompted the current study to investigate the factors that contribute to the contradictory results in Botswana. Theoretically, Botswana appears as a dual economy (Hirschman 1958) when in practice, it is not. The rural economy is linked to the modern economy through government spending policy, migrant remittances, and the rural development policy. Therefore, rural economy in Botswana benefits directly from sustained economic growth.
The cost of basic needs method and the mean income poverty line are applied in this thesis. Two subjective measures of poverty were adopted from Firdausy and Tisdell (1992), and Pradhan and Ravallion (2000) to explore whether they might be reliable low cost rural poverty indicators. In addition, economic deprivation is measured and results provide a basis to examining how future policies of redistribution can influence poverty reduction. The results are based on the case study of Nshakazhogwe village in rural north east Botswana of September-November 2005. This village was purposively chosen as a representative cluster and a face-to-face personal interview of all heads of household data collection method was used. The results show that women are overrepresented as heads of household which is common in Botswana (Quisumbing et al. 2001; Rosenhouse 1989). Botswana has an equal opportunity policy across gender (Ministry of Labour and Home Affairs 1995) and it has been found that female heads of households are common in other countries where opportunities to education and work for women are similar to men’s (Schultz 2001). The results indicate that the village characteristics are comparable to those of the rural north east Botswana region (Central Statistics Office 2007; Central Statistics Office Botswana 2001a; Central Statistics Office Botswana 2004), which means this village is typical for the region.
The incidence of poverty at Nshakazhogwe is 29 per cent, which is lower than 45 per cent for all rural areas. The Lorenz curve analysis shows that the income share of the bottom 20 per cent share of the population is 2.7 per cent whereas the income share of the top 20 per cent share of the population is 65 per cent. The income inequality measure using the Gini index is 60 per cent which is comparable to Botswana Gini index in 2002/03 but less than the Gini index for the whole of Africa or the global Gini index. The results of application of subjective question of whether household heads believe their household belongs to the bottom 20 per cent less well-off household reveal that 42 per cent households reported positively, implying they were in poverty, which is a clear overestimate. However, the result is comparable to national rural poverty of 45 per cent using the mean income method (Central Statistics Office Botswana 2004). The results using a consumption adequacy method is 37 per cent, which is comparable to the rural north east region poverty rate (Central Statistics Office 2007). Overall, subjective measures used in this study are reliable indicators in distinguishing the well-off from the less well-off. The significance of this finding is that, if applied in rural Botswana, it might contribute to quicker, relevant and appropriate responses to poverty rural problems. Further research is required to verify the results of this study. Income transfers are very important at Nshakazhogwe. More poor households than non-poor households receive private transfers, whereas the relative frequency of poor households that receive government transfers is comparable to not poor households. Livestock, natural resource harvesting and crop production are not indicated frequently as sources of household income. Future research should investigate the target efficiency of government transfers and the relative benefits of individuals from each source of income in rural areas. Potential policy implications of this study are an increase in availability and regularity of reliable poverty indicators and an improvement of targeting efficiency of government welfare.