In this paper, I empirically investigate precautionary savings under liquidity constraints in Italy using household panel data. Following Jappelli and Pistaferri (2000) I analyze a 3- year (1989-1993) rotating panel of the Bank of Italy Survey of Household Income and Wealth (SHIW). I exploit a unique indicator of subjective variance of income growth, which allows to measure the strength of the precautionary motive for saving, and a variety of survey-based indicators of liquidity constraints. However, my analysis deviates from Jappelli and Pistaferri’s in three aspects. First of all, I attempt to differentiate between the standard precautionary motive for saving caused by uncertainty from the one due to liquidity constraints. I address this issue by using an endogenous switching regression approach, which allows me to cope with endogeneity issues associated with sample splitting techniques. Secondly, I try to capture changes in consumption behaviour of households who are not constrained at present , but expect binding constraints in the future. Finally,I cope with the downward bias in the estimation of the parameter associated to the subjective variance of income growth, using a direct measure of risk aversion. I eventually found the precautionary motive for savings to be stronger for those households who face binding constraints, or expect constraints to be binding in the future. Indeed, a complementarity relation exists between precautionary savings and liquidity constraints. Moreover, the introduction of a survey-based measure of risk aversion allows a better identification of the coefficient associated with the subjective measure of variance of income growth.