Demand for large and indivisible, or “lumpy”, expenditures creates a need for liq- uidity. To acquire this liquidity, people in developing countries are often forced to choose among high-cost strategies. I conduct a study with 1,715 bettors in Kampala, Uganda, to show that sports betting is being used as an alternative to conventional liquidity generation strategies such as saving or credit. First, I document that, despite expected losses of 35-50%, participants view betting as a likely source of liquidity for desired lumpy expenditures and use a natural experiment to show that this is not just cheap talk: winnings increase both the size and likelihood of making such expendi- tures. Second, I use a randomized field experiment to show that provision of a simple commitment-savings technology causes a 26% reduction in a revealed preference mea- sure of betting demand. I then conduct two lab-in-the-field experiments to isolate the role of betting as a mode of liquidity generation. Increasing the salience of a desired lumpy expenditure causes an increase in betting demand by 17.2%, and a budgeting exercise decreases betting demand by 34.9% for people who learn that they could save more than previously believed. Back-of-the-envelope calculations suggest that betting to create liquidity may be a rational response for people with low ability to save.