Using three major UK pension reforms as natural experiments we investigate the relationship between pension saving and discretionary private savings. Unlike most differences-in-differences approaches which rely on average differences between control and treatment group, we use economic theory to model the response of each individual household. The empirical analysis, based on the Family Expenditure Survey, uses both time-series and cross-sectional variation to identify the behavioral response. The earnings-related tier of the pension scheme is found to have a negative impact on private savings with relatively high substitution elasticities; the impact of the flat-rate tier is not significantly different from zero.