This paper quantifies the impacts of policy objectives on the composition of an optimum new passenger vehicle fleet. The objectives are to reduce individually absolute energy use and associated emissions of CO2, NOx and PM2.5. This work combines a top down, diversity-led approach to fleet composition with bottom-up models of 23 powertrain variants across nine vehicle segments. Changing the annual distance travelled only led to the smallest change in fleet composition because driving less mitigated the need to shift to smaller vehicles or more efficient powertrains. Instead, managing activity led to a 're-petrolisation' of the fleet which yielded the largest reductions in emissions of NOx and PM2.5. The hybrid approach of changing annual distance travelled and increasing willingness to accept longer payback times incorporates management of vehicle activity with consumers' demand for novel vehicle powertrains. Combining these changes in behaviour, without feebates, allowed the hybrid approach to return the largest reductions in energy use and CO2 emissions. Introducing feebates makes low emitting vehicles more affordable and represents a supply side push for novel powertrains. The largest reductions in energy use and associated emissions occurred without any consumer behaviour change, but required large fees ((sic)79-99 per g CO2/km) on high-emitting vehicles and were achieved using the most specialised fleets. However, such fleets may not present consumers with sufficient choice to be attractive. The fleet with best diversity by vehicle size and powertrain type was achieved with both the external incentive of the feebate and consumers modifying their activity. This work has a number of potential audiences: governments and policy makers may use the framework to understand how to accommodate the growth in vehicle use with pledged reductions in emissions; and original equipment manufacturers may take advantage of the bottom-up, vehicle powertrain inputs to understand the role their technology can play in a fleet under the influence of consumer behaviour change, external incentives and policy objectives. (C) 2016 Elsevier Ltd. All rights reserved.