Background: Teacher quality matters for student achievement (Coleman, 1968; Rivkin et al., 2005; Rockoff, 2004; Aaronson et al., 2007) and later career success (Chetty et al., 2014). States use licensure exams as a quality screen believing that they are predictive of teaching effectiveness (Council et al., 2001). However, the evidence on the relationship between licensure exam scores and students' success is mixed and often faces a variety of data limitations. Goldhaber, D. (2007) found that the students whose teachers' scores were at the top quintile in curriculum tests scored 2.2% standard deviation higher in reading and 3.5% standard deviation higher in math than those whose teachers' scores were at the bottom quintile. Cowan et al. (2020) estimated that one standard deviation increase in teachers' Massachusetts Tests for Educator Licensure (MTEL) performance is associated with an improvement of 2.4% standard deviations in student math test scores. However, Buddin & Zamarro (2009) found that teacher licensure exam scores were unrelated to student achievement. Likewise, Chung & Zou (2021) found that Educative Teacher Performance Assessment (edTPA) did not predict student test scores well. Thus, more evidence is needed on the effectiveness of licensure exams as a teacher quality screen. Licensure exams limit the supply of certified teachers, and if these exams are an ineffective quality screen, then supply-improving changes might be made without sacrificing teacher quality. Teacher shortages are a longstanding challenge in many states, especially in certain geographic areas and subjects like mathematics, science, and special education. COVID-19 may have meaningfully increased teacher turnover in the subsequent years (Camp et al., 2023), especially among new teachers (Bacher-Hicks et al., 2023), which has the potential to exacerbate existing shortages. Moreover, licensure tests disproportionately negatively affect teachers of color. For example, Cowan et. al. (2023) observed that Black and Hispanic teacher candidates have lower pass rates and retake exams at lower rates than White candidates. Due to these pressures, many states relaxed or waived licensure exam requirements during the pandemic (Will, 2022), and several states have or are considering permanent changes to their licensure exam regiments and cut scores. Better understanding the relationship between licensure exam scores and classroom effectiveness is highly important in this evolving policy environment. Research Questions: We provide evidence on the relationship between licensure exam scores and teaching effectiveness using data from Arkansas, a state that has recently made changes to its licensure exam cut scores and has several licensure waiver programs. Specifically, we address the following research questions: 1. Do licensure test scores, particularly cut scores, are working as an effective screening of licensure and employment as teacher? 2. What is the impact of passing/not passing a teacher licensure exam on teaching effectiveness as measured by value added (VAM)? Data Sources: We use individual-level licensure exam scores, employment, job-assignment, and value-added data covering all licensure exam test takers and public-school employees from 2015 through 2023. These data are maintained by the Office of Education Policy at the University of Arkansas. We combine these data with detailed individual enrollment and graduation records from the Arkansas Department of Higher Education, which includes the universe of public and private higher education institutions in the state. Additionally, we link these individual records with district and school characteristics from the Common Core of Data and Arkansas Department of Education's Data Center. Methods: We employ quasi-experimental methods, specifically regression discontinuity design (RDD) for our analysis. In our estimation, we utilized RDD models using distance from the state cut score as the running variable and passing/not passing on the first attempt of licensure exam as treatment variable. We estimate models using local linear and second-order polynomial regression with MSE-optimal bandwidth using the heteroskedastic-robust nearest neighbor variance estimator. Additionally, we control for gender, race/ethnicity, EPP program type, and school level in our analysis. Preliminary findings: Our preliminary analysis shows that in Arkansas Licensure exams do not appear to be effectively deterring people from getting a license or becoming employed as a teacher. These exams create temporary hurdles and increase costs for aspiring teachers. Moreover, teachers' value added (VAM) appears quite similar for the teachers who passed and did not pass licensure exam at the first attempt. Thus, licensure exam cut scores could potentially be reduced to eliminate unnecessary barriers and costs.