Population projections are key for policy making and planning. Currently, most countries, and international organisations like the World Bank and the United Nations, produce population projections using the cohort-component method. The method is simple and applicable in absence of detailed data sources. But it is limited to projecting a population by age, sex, and very few additional variables, such as province or broad education categories. A more advanced – but still uncommon – approach involves dynamic micro-simulation models, where populations are represented by large samples of individual people and their life-courses are simulated over time. This approach is more complex, but has major advantages: it can produce detailed projections of a broad variety of individual characteristics, it can model realistic life-courses and their diversity, and it supports the modelling of interactions between people. Such models also support more detailed planning and policy development and can provide the demographic core of more extensive socioeconomic models. While currently applied almost exclusively in the developed world, the benefits can be highly significant in the context of developing countries.