Given the importance of reliable recruitment estimates when assessing temperate eel stocks and enforcing appropriate management measures, surprisingly few analytical tools have been developed to estimate yearly glass eel recruitment. Of the models that do exist, large-scale models generally rely on strong assumptions relating to fishing activity, while other models generally estimate recruitment at the river basin scale. With the aim of filling this gap, we developed the GEREM (Glass Eel Recruitment Estimation Model) to estimate glass eel recruitment at different nested spatial scales. Our model simultaneously estimates annual recruitment at the river catchment level, at an intermediate spatial scale such as Eel Management Units (EMUs), and at a larger scale (e.g., a country). Provided sufficient data become available in the future, the analysis could be extended to the scale of the distribution area, which would be consistent with the population scale. In this study, the model was applied to France, using various recruitment indices obtained from 1960 to 2013. This provided trends and absolute recruitment estimates consistent with current expert knowledge. A sensitivity analysis was carried out to assess the robustness of results to sources of uncertainty. This type of model fills an important gap in the range of quantitative tools presently available to estimate recruitment. It could be used in the future to establish total allowable catches in countries such as France where glass eels are fished commercially.