Demographic ageing in New Zealand has greatly increased the proportion of older people, with major implications for the provision of health and social care. Policy options include promoting healthier ageing, and changing the balance of care. To test these options, we first constructed a micro-simulation model of the 65+ life course. This type of modelling relies heavily on the availability of micro-level data. Often, only official sources can provide such data without further costly and time-consuming collection. We combined data from two ongoing official national survey series, on health and disability respectively, to generate a synthetic version that replicated original data and parameter settings. These official data were particularly advantageous as they were derived from nationally representative samples so that results from modelling could be generalised to the New Zealand population in the future. However, there were methodological challenges related to: data harmonisation across years within a series, and between the two series; data imputation to fill gaps; the calculation of transition probabilities to drive the simulation in a dynamic way; and the scope and detail of extant variables being sufficient to model the core processes and key outcomes of substantive interest. In the building of the model, the official data were used for four purposes: to form a starting sample of older people at the base year; to derive statistical equations to inform cross-sectional dependencies and outcomes in any year; to derive transition probabilities to inform dynamic changes in future years; and to provide benchmarks for replication. We then used the final model to artificially modify morbidity levels (long-term illness and disability) or the balance of care across a range of modalities (in health services, and in community and residential care), and to observe the impact on the overall use of care. We report on both the construction of the model and the results of simulated policy-relevant scenarios. We conclude that, at a technical level, existing official data sources can be successfully used for modelling purposes; in fact, they may be crucial to that success in the absence of other data. At a substantive level, we conclude that demographic ageing may not have a major negative impact on system resources, while changing the balance of care may make better use of those resources.