Existing methods for the assessment of the potential impacts of climate change in productive activities and sectors are usually limited to point estimates that do not consider the inherent variability and uncertainty of climatic and socioeconomic variables. This is a major drawback given that only a limited and potentially misleading estimation of risk can be expected when ignoring such determinant factors. In this paper, a new methodology is introduced that is capable of integrating the agent's beliefs and expert judgment into the assessment of the potential impacts of climate change in a quantitative manner by means of an objective procedure. The goal is to produce tailor-made information to assist decision-making under uncertainty in a way that is consistent with the current state of knowledge and the available subjective "expert" information. Time-charts of the evolution of different risk measures, that can be relevant for assisting decision-making and planning, can be constructed using this new methodology. This methodology is illustrated with a case study of coffee production in Mexico. Time-dependent probabilistic scenarios for coffee production and income, conditional on the agent's beliefs and expert judgment, are developed for the average producer under uncertain future conditions. It is shown that variability in production and income, generated by introducing climate variability and uncertainty are important factors affecting decision-making and the assessment of economic viability that are frequently ignored. The concept of Value at Risk, commonly applied in financial risk management, is introduced as a means for estimating the maximum expected loss for a previously chosen confidence level. Results are tailor-made for agents that have incomplete information and different beliefs. In this case study, the costs of climate change for coffee production in Veracruz are estimated to have a present value representing from 3 to 14 times the current annual value of coffee production in the state. © 2011 The Author(s).