This paper focuses on the role of habit formation in individual preferencesover consumption and saving. We closely relate to Alessie and Lusardi's(1997) model as we estimate a model which is based on their closed-formsolution, where saving is expressed as a function of lagged saving and otherregressors. Alternatively, we could use an Euler-equation approach (see e.g.Guariglia and Rossi (2001) and Dynan (2000)), but we will argue that thisapproach may yield spuriously negative estimates of the habit formationparameter because in surveys consumption is typically measured withconsiderable error. A second reason to use the closed form solution as abasis of the empirical model is that it embodies more information about thehabit formation model than the Euler equation. Therefore, the closed formsolution allows for a more powerful test of the validity of the habitformation model than the Euler equation approach.