Credit market imperfections make households more vulnerable to shocks and their consumption decisions extremely sensitive the timing of their income. Algorithmic insights from individual meta data have enabled a proliferation of mobile financial services to cellphone users in markets where information asymmetries and high provision costs tend to deter formal financial institutions. As the first such financial product typically offered to new users, airtime loans provide prepaid customers small airtime advances for a fee as an alternative to recharges purchased from network agents. Relying on rich administrative data from a mobile network operator in Haiti, we study the impact of airtime loans on consumer cellphone expenditure and network usage. We find that access to loans increases total communication expenditure by 16% due to a crowding-in of additional network usage. This expenditure response to airtime loans is distinctly heterogeneous. Poorer customers in the lowest tercile of initial expenditure more than double their mobile communication spending when airtime loans become available, while access to loans leaves expenditure of the highest tercile unchanged. These differences in the expenditure impacts of airtime loans exist despite relatively uniform patterns of loan usages between the poor and non-poor. We find suggestive evidence that these differences are driven by distinct motivations for requesting airtime loans $:$ Poorer customers appear to use loans to relax liquidity constraints at critical communication times whereas non-poor customers primarily use these loans for convenience, as it gives them more discretion in when to visit airtime vendors.