We quantified the relative importance of the precautionary motive in determining savings. Existing empirical evidence suggests that the impact of precautionary savings is small if one uses a subjective (based on self-reported expectations) measure of uncertainty about next year income. However, other studies use more objective (based on income data) methods to proxy for income uncertainty by exploiting life-cycle income variation. These studies find a large effect of precautionary savings. These contradictory results may be either suggestive of methodological shortcomings or the result of institutional differences between countries. We refined the subjective method by taking into account the uncertainty as perceived by the second income earner. We then apply the 'objective' and 'subjective' method to the same data set and obtain similar results: the subjective and objective methods suggest that precautionary savings account for approximately 30% of savings. © 2013 Oxford University Press. All rights reserved.