This paper introduces a methodological innovation into Generational Accounting. By incorporating cyclically-adjusted balances into the forward-looking budget projections underlying the concept we isolate pure policy effects, which render comparisons of the fiscal sustainability indicators obtained across time and countries truly meaningful. We also show that a demographic effect and a debt effect may bias fiscal sustainability measures over time, and establish a routine to control for these effects in the generational accounting framework. An empirical application for Spain illustrates that our proposed decomposition of indicators is empirically relevant. Standard generational accounting suggests that fiscal sustainability in Spain improved substantially in preparing for EMU. However, calculation of the pure policy effects reveals that this has not been the case.