If scarce community resources for emergency services are to be allocated efficiently from a social perspective, decision-makers should be informed by social cost-benefit analysis (CBA). However, this is rarely, if ever, the case in Australia. One reason may be the challenge of estimating benefits that are not provided fully by private markets. In a case study of post-cyclone emergency services in Cairns, Australia, we employ Choice Modelling to estimate the willingness to pay of households for a set of publicly supplied emergency services. Accelerated reconnection of utilities like sewerage and electricity is the most highly valued, about three times as much as faster resupply of fresh food, while additional police patrols are far less valued and accommodation for pets negatively so. These results are reflected in a standard, deterministic CBA that suggests that all but pet accommodation should be publicly provided. However, Monte Carlo analysis incorporating variability in estimates of costs and benefits resulted in lower mean expected levels of Net Present Values, a salutary reminder that deterministic estimates may not be appropriate in circumstances of uncertainty about costs and benefits.