Skip to main
University-wide Navigation

Racial Inequities in Effective Food Sales Tax Burdens

Samanta Laskar

Abstract: Food sales tax policies in the United States vary widely across states and across food categories. Some states fully tax groceries, others tax them at reduced rates, and many exempt groceries altogether. At the same time, even in states that exempt groceries, prepared foods, soda, and candy are generally taxed at the full sales tax rate. As a result, households with may face different effective food sales tax burdens depending on where they live and what types of food they purchase. This paper examines whether these place-based and product-specific tax rules generate unequal effective food sales tax burdens across Black and White households depending on where they live and what they buy.

Heterogeneous Returns to Sub-Baccalaureate Credentials: Machine Learning Approach

Kangwook Noh

Abstract: This study examines the labor market returns to sub-baccalaureate credentials, such as associate degrees and certificates, using comprehensive administrative data from Kentucky’s postsecondary education system. By linking over a decade of student records to quarterly earnings data from the state’s Unemployment Insurance (UI) program, we provide a detailed analysis of labor market outcomes for credential completers versus non-completers. Building on previous research (e.g., Jepsen, Troske, and Coomes, 2014; Carruthers and Sanford, 2018), which often employs Difference-in-Differences designs with individual fixed effects, our study contributes to the literature by employing Double Machine Learning (DML) methods. Specifically, we estimate Conditional Average Treatment Effects to capture complex, non-linear patterns of heterogeneity in returns, showing how outcomes differ by  demographics and pre-program labor market experience. Additionally, we produce more precise “apple-to-apple” comparisons and estimate the treatment effects of completing sub-baccalaureate credentials. We are unable to report the findings until going through export review by the data partner (which will be completed by the time of the conference). However, our preliminary results aligns with prior literature in showing significant earnings increase for students who earned sub-baccalaureate credentials, compared to those who pursued but did not complete the program.