Several recent IFPRI studies have measured the effects of public spending on growth and poverty reduction in selected Asian countries using pooled time-series and cross-region data. However, many African countries lack such data. Using Tanzania as a test case, this study demonstrates how household survey data can be used to assess the impacts of public investments on growth and poverty. A two step procedure is used. First, household survey data are used to link household welfare measures to human capital and household access to infrastructure and technology, while controlling for other community and household characteristics. The second step links household human capital and access to infrastructure and technology to past public investments in these factors. As in the Asian studies, the growth effects (measured as per capita income) of investments in agricultural research, roads, and education are found to be large. But unlike Asia, no clear distinction emerges between the measured impacts for high and low potential areas. In many high potential areas, returns to investments are still high and there is no sign of any diminishing marginal returns. This suggests that there has been insufficient public investment in all kinds of regions. Nevertheless, the results show that there is opportunity to improve on the growth and poverty impacts of total public investment through better regional targeting of specific types of investment. For example, additional investments in rural education have attractive growth and poverty impacts in all regions, whereas additional investments in roads and agricultural research are better spent in the central and southern regions of the country.