Background/Context: Both correlational and experimental evidence suggests that having access to books can be a key resource for accelerating the literacy outcomes for children. Results from a 20-year study across 27 nations by Evans, Kelley, Sikora, and Treiman (2010), suggest that children growing up in homes with many books obtain 3 years more schooling than children from bookless homes, independent of their parents' education, occupation, and social class. Yet, many families in the United States have few or no children's books in the home. In some of the poorest communities, books are simply unavailable. For example, Neuman and Celano (2001) found a ratio of one book for every 300 children in low-income Philadelphia neighborhoods. Since 2009, Bernie's Book Bank has established an effective and replicable model for collecting millions of new and gently used surplus books and delivering them directly into the hands of low-income infants, toddlers, and school-aged children who have few or none of their own. To date, Bernie's Book Bank has provided over 7 million books to economically disadvantaged students across the Chicagoland area. Currently, more than 250,000 children in Chicago and its suburbs receive a minimum of 12 free books per year through the Book Bank. On the face of it, these efforts seem to be helping close the book-ownership gap between low-income students and their relatively higher-income peers. Purpose/Research Question: This study answers the research question: Does school-level assignment to the Bernie's Book Bank program impact the reading achievement outcomes of treated students and schools relative to a business as usual control group? Setting: This study included students in 60 elementary schools in Milwaukee, Wisconsin. Participants: The analytic sample for these analyses includes 2,142 students who were in grade 1 or 2 in 2018-19 (baseline). Nearly all students were classified as eligible for free/reduced price lunch (96%), approximately 85% were Black or Latinx, 26% were eligible for special education services and 11% were English learners. Intervention: Students in the treatment schools received six books during the fall of the school year and another six books during the spring of the school year across five years. Research Design: In this school-level random-assignment design, participating elementary schools randomly assigned to treatment received free books for each student in grades K-6 while students from schools assigned to the control condition received the free books only after the completion of the study. Students received 6-12 books per year for the five years. Data Collection and Analysis: Measures: The grade 1 and 2 students' scores on the fall 2018-19 STAR Early Literacy and Reading assessment serve as a baseline. For the Year 5 literacy outcome measure, we use the 2022-23 state literacy assessment, FORWARD. We standardize all pretests and posttests relative to the districtwide Milwaukee means and standard deviations for each specific test and grade level. This standardization provides an accepted methodology that enables us to combine test scores across grade levels and assessments in a single analysis (May et al., 2009). We also include student-level demographics: race/ethnicity, gender, free/reduced-price lunch status, English language learner status, and special education status. The primary intention-to-treat (ITT) research question is addressed using the following hierarchical linear model: [equation omitted]. The level-one, or within-school model, nests the grade students within schools with the spring 2022-23 FORWARD outcome for student i in school j, Y[subscript ij], predicted by a school-level intercept, [beta][subscript 0j], a vector of student-level covariates, including the students' STAR pretest score, baseline grade level (1 or 2), race/ethnicity, gender, free/reduced-price lunch status, English language learner status, and special education status, and a student-specific level-one residual variance component, r[subscript ij]. At level 2, the school-level intercept, [beta][subscript 0j], is predicted by a school-level grand mean, [gamma][subscript 00], the cluster-level impact of Bernie's Book Bank treatment assignment on the mean posttest achievement outcome in school j, plus a school-level residual, u[subscript 0j]. Findings/Results: The analytic sample for these analyses includes 2,142 students who were in grade 1 or 2 in fall 2018-19 (baseline). At the school-level, all 30 control schools and 30 treatment schools are included in the analysis. As shown in Table 1, there were no statistically significant treatment-control differences for any of the baseline mean school-level covariates. Further, data attrition claimed a comparable share of the student sample in treatment (58%) and control sites (59%), which is not statistically different at a conventional level. Table 2 shows a combined overall student-level attrition rate of 59% and a differential attrition rate of 0.80 percentage points. Relative to the What Works Clearinghouse's (WWC) most recent standards, our attrition results meet standards under the "optimistic boundary" (WWC, 2022). Our ITT analyses include all students as they were originally randomized. Results show that there is a statistically significant impact of assignment to treatment on the spring 2022-23 FORWARD reading achievement scores. The results for the baseline grade 1 and 2 sample show a point estimate equivalent to an effect size of approximately d = 0.10, indicating that treatment schools had statistically significantly higher posttest achievement scores (p < 0.05, [beta] = 0.096) (see Table 3). Conclusions: A recent national survey by Scholastic (2015) suggests that across all levels of family income, 9 of 10 children ages 6-17 report that they enjoy reading books for fun. However, about 60% of children who live in the lowest-income households report that they rely mainly on their schools for the books that they read for fun, while only about 33% of children ages 6-17 who live in the highest-income homes indicate the same. This evidence suggests that all children enjoy reading for fun, but poor children have limited access to books in their homes and communities and are, therefore, more reliant on their schools to provide books and reading opportunities than are more affluent students. Bernie's Book Bank provides an easily replicable and cost-effective strategy to combat the "book deserts" found in so many low-income schools and communities. Our results suggest that helping poor students develop their home libraries can positively impact their high-stakes reading outcomes.