The paper examines the dynamic relationship between the series of monetary aggregates M1 and M2 for the period 1972-2014. M1 and M2 are the dependent variables, while the explanatory variables are real income, discount rate, inflation rate, real exchange rate, and remittances. The ARDL bounds testing approach to cointegration is used to investigate the existence of long-run and short-run effects of remittances on monetary aggregates. The results show that remittances exert only positive effects on real narrow money demand in the end, suggesting that in Pakistan remittances are used for the purpose of consumption. Both money demand functions are stable in Pakistan, but the longrun effect of M1 remittances is a faster speed of adjustment to equilibrium (26.2%) than M2 remittances (21.3%). It is recommended that M1 be used as a monetary tool in Pakistan.