Since 2006, Mali has experienced the full effects of the global food crisis, with price increases of up to 67%. This study presents simulations of the impacts of this crisis and a number of policy responses with respect to the welfare of children. The impacts are analyzed in terms of monetary (food) poverty, nutrition, education, child labour and access to health services of children. According to simulations, food poverty among children would have increased from 41% to 51%, with a corresponding rise in caloric insufficiency from 32% to 40%, while the impacts on school participation, work and access to health services would have been relatively weak. To prepare an adequate response, the government should start by identifying the poor individuals who are to be protected, based on a limited number of easily observed sociodemographic characteristics. A method of targeting these individuals is proposed in this study. However, simulations show that with targeting about one quarter of poor children would be erroneously excluded (under-coverage), while more than a third of non-poor children would be erroneously included (leakage). These identification errors, which increase in proportion with the extremity of poverty, reduce the impact and increase the cost of any public interventions. That having been said, it is important to note that leakage to the non-poor can nonetheless improve the conditions of children in terms of caloric intake, school participation, child labour and access to health services, none of which are exclusive to poor children. When targeting children or sub-groups of children by age, benefits will likely be deflected to some extent to other family members. Moreover, it is total household income, regardless of the member targeted, that determines decisions relating to child work, education or access to health services. School feeding programmes are found to be a particularly efficient policy in that they concentrate public funds exclusively on the consumption of highly nutritious foods, while cash transfers can be used by households for other purposes. Moreover, school feeding programmes are likely to have desirable effects on school participation and child labour. However, there are some caveats due to the fact that these programmes exclude children who do not attend school, the difficulty of exclusively targeting poor children and the possibility that child food rations at home will be proportionally reduced.