As college promise programs proliferate across the United States with noted intentions to promote access through increased affordability, it is necessary to understand the relationship between these programs and other forms of financial aid, including loans. Using federal, state, and program-level data, we leverage a natural experiment to estimate causal impacts of the nation's first statewide program (Tennessee Promise) on students' borrowing behaviors in a community college system. In the difference-in-differences framework, we find robust evidence to suggest Tennessee Promise reduced the percent of first-time, full-time students borrowing by 8-10 percentage points on average after implementation--an over 40% decline--and reduced the average community college cohort loan by $230-360 (nearly 32%). While consistent with prior work on grant aid, these findings are among the first to explore promise students' financial outcomes. We discuss these results in relation to Tennessee Promise's programmatic features and identify important implications for policy and research. [This paper will be published in "The Journal of Higher Education."]