Full text can be accessed https://www.jstor.org/stable/23026305This paper examines the hypothesis that dependency rates and poverty adversely affect private savings in developing countries (LDCs) by using panel data from a one-shot survey of peasant households in Northern Tanzania. Overall, the multivariate regression results demonstrate that dependency rates have a statistically significant influence on savings in poor households and an insignificant positive effect on savings in rich households. The same positive but insignificant influence of dependency rates also obtains in the overall sample results and, by and large, when the dependency ratio is disaggregated. Empirical analysis also shows that wealth and income are determinants of savings in the rich peasant households and only cultivated land (per capita) determines savings in poor peasant households. The estimated income elasticity of savings is greater in the rich household than in the poor households, implying the latter saves less out of current income. The empirical results of this paper fail to support the 'claimed' adverse effect of dependency rates on savings in the LDCs, at least in the sampled area in Tanzania. This empirical finding undermines birth control policy measures in Tanzania. Besides, to the extent that rich households save more out of income and wealth underscores targeting of financial saving instruments to the rich rural households and, the adoption of policies that would increase saving capacities of poor households is implicit from the results. The findings and the main conclusions of this study need, however, to be treated cautiously because of the specificity of the area covered. Further exposure to data of the dependency ratios-savings nexus to household budget survey data is required