In this paper we examine the statistical properties of the monetary aggregates in Chile in order to know if the time series display mean reverting behaviour. For this purpose, we use techniques based on fractional integration. Monthly data of various Chilean monetary aggregates from January 1986 until August 2019 are used, and the results indicate very weak evidence of mean reversion. In fact, this property is only found in the case of the currency on circulation and M1 for some of the series examined; however, for M2 and M3 the results clearly show high persistence with orders of integration substantially higher than 1. Thus, shocks are expected to have a permanent nature in these cases. Another remarkable feature observed in the results is that the level of persistence in the series seems to grow with the amount of the monetary aggregate. In a multivariate context, performing a FCVAR model, evidence of cointegration is found among the monetary aggregates, finding a long run equilibrium relationship between them.