The use of climate information in economic activities, typically provided by climate services, may serve as a possible adaptation strategy to changing climate conditions. The present paper analyzes the value of climate services aimed at improving agricultural productivity through a reduction in weather-associated risks. In the first part, we provide a theoretical foundation for estimating the value of climate services by proposing a stochastic life-cycle model of a rural household which faces uncertainty with respect to the timing and the size of an adverse weather shock. We subsequently calibrate the model to match the environment of coffee producers in the Cusco region of Peru and provide a range of estimates for the value of climate services for a single average household, the region, and the country as a whole. In the second part of the paper we use empirical data to verify the numerical estimates. We assess the value of climate services in the agricultural sector in Cusco based on a choice experiment approach. Data are analyzed using a standard as well as a random parameter logit model allowing for preference heterogeneity. Farmers show a significant willingness-to-pay for enhanced climate services which is particularly related to the service accuracy and geographic resolution. On average, the yearly value of a climate service in the coffee sector is found to be in the range $20.64 - $21.10 per hectare and $8.1 - $8.2 million for Peru as a whole.