Intergenerational earnings mobility studies often overlook changes in earnings variance over a person's life. By analyzing data from Canada and the US, researchers found that estimated earnings persistence is strongly influenced by the age at which data is collected. This age-dependence is partly due to an increase in temporary earnings fluctuations during data collection. Additionally, the study identified an independent effect of life cycle investment on earnings persistence. Variations in fathers' ages among studies contribute to one-third of the differences in published estimates of earnings persistence. These findings challenge the validity of tests for credit constraints based on earnings at different life stages.