The ‘new variant famine’ (NVF) hypothesis postulates that the HIV/AIDS pandemic is eroding rural livelihoods and making rural households more vulnerable to drought and other transitory shocks. Despite limited empirical evidence, the NVF hypothesis has become an important part of the conventional wisdom surrounding the relationship between HIV/AIDS and food crises in southern Africa. This study provides a new empirical test of the NVF hypothesis via econometric estimation of the relationship between AIDSrelated morbidity and mortality and indicators of rural livelihoods. District longitudinal data from smallholder farmers in Zambia surveyed annually between 1991 and 2003 are used to estimate several econometric models in order to: (1) understand the effects of HIV/AIDS on rural farm production; (2) measure whether HIV/AIDS exacerbates the impacts of drought and other factors affecting rural farm production; and (3) determine whether these results are consistent with the predictions of the NVF hypothesis. We find little evidence of a systematic decline in rural livelihoods at the national or provincial level as measured by mean household agricultural production, area cultivated, or the value of production per unit of land. Furthermore, contrary to a priori expectations, we do not find evidence of a robust negative direct effect of HIV/AIDS on any of these three agricultural production outcomes. We do find some evidence that HIV/AIDS may have negative indirect effects on rural farm production by exacerbating the impacts of drought, gender inequalities and agricultural sector policy changes related to structural adjustment. This final finding is consistent with the predictions of the NVF hypothesis.