Despite increases in college enrollment, college completion rates have not kept up, at both 2-year and 4-year institutions. In California specifically, the 6-year graduation or transfer rate for students at 2-year institutions was only 48%. In response to the idea that students arrive at college unprepared, driving these low college success rates, colleges have implemented a system of remedial education to address these concerns. Remedial education is meant to help prepare students for the rigors of college courses and do not confer any college credit to the student. Importantly, students aren't allowed to take college-level, credit-bearing college courses if they haven't completed the corresponding remedial education sequence. Descriptive studies have shown that remedial education can have unintentional negative consequences. In California community colleges specifically, remedial education has served to elongate time to degree and encourage attrition. Remediation sequences are lengthy, with English remediation sequences taking, on average, 1.9 terms for completion, and math remediation requiring an average of 2.5 terms for completion. Furthermore, attrition is extremely high, with only 44% of developmental math students and 60% of developmental English students completing the sequence. Previous empirical literature regarding the effectiveness of remedial education is mixed. Some papers find that remedial education had positive effects on short-run outcomes, such as college persistence and transferring to a more selective college. Others have found, however, no effects on degree completion or labor market outcomes. As a result of these studies, Assembly Bill 705 (AB 705) was passed in October 2017 in California, addressing these problems, and changing the way schools placed students into remedial education, with mandatory implementation by Fall 2019. Colleges must "maximize the probability that a student will enter and complete transfer-level coursework in English and math within a one-year timeframe." Consequently, more students can directly enroll in transfer-level courses, decreasing time to degree or transfer by allowing students to take classes more relevant and more efficiently. Alternatively, if these students really were underprepared and required remedial education, then allowing direct access to college-level courses might result in additional course failure and attrition. I use the introduction of AB 705 as a source of quasi-exogenous variation to study how open access to college-level courses affect students' college success in California Community Colleges, including short-run measures, such as grades in transfer-level courses, overall GPA, as well as medium-run measures, such as persistence, and 2-year graduation rates. I implement a difference-in-differences framework, comparing college outcomes for students before and after Fall 2018, and across different levels of predicted treatment status. There are a few wrinkles. Although AB 705 mandated that California community colleges apply these policies by Fall 2019, some schools decided to pilot these programs early in Fall 2018. Consequently, I define the post period to include and be after Fall 2018. Next, I choose demographic, socioeconomic, and ability characteristics using the LASSO method to find a model to predict the probability that a student would have been placed in remedial education, projecting counterfactual treatment status. I can implement this identification strategy by taking advantage of rich administrative K-12 educational data at the student level from the California Department of Education from 2000 -- 2019. Importantly, this data allows me to control for student ability, using 11 grade standardized test scores, as well as socioeconomic and demographic variables, like free-lunch status, race, and gender. I then link the K-12 data to equally as rich and comprehensive student-level California community college administrative data over the same period. The community college data includes exhaustive information on courses, including remedial education status, as well as number of students enrolled, by course and by school. Furthermore, the community college data is at the student-course level, and includes grades at the course level, units attempted, units earned, and course subject. This extensive community college data will allow me to control for any potential selection issues, such as student ability, but also any selection issues at the college level, for those colleges who decided to select into reducing their remedial education course offerings early. I find that, overall, AB 705 had a strong first stage, with remedial English and math participation decreasing, and corresponding transfer English and math participation increasing. That is, a one standard deviation increase of 20.5% in the predicted probability of enrolling in remedial English leads to a 12.5% decline in the probability of enrolling in remedial English, while a one standard deviation increase in of 20.1% in the predicted probability of enrolling in remedial math leads to a 11.2% decline in the probability of enrolling in remedial math. Although there is a corresponding increase in the probability of enrolling in transfer English and math, the increases are not one-to-one, suggesting that there is still some hesitation among students who would have had to take remedial education in enrolling in transfer-level courses. However, there is still an increase in the probability of those passing transfer-level courses with a C- or better in both English and math. However, these positive effects mask heterogeneous responses by students, depending on their predicted probability of enrolling in remedial education. For example, although students in all quartiles reduced their remedial education participation, the corresponding increase in transfer level participation decreased as the probability of remedial education participation increased. Similarly, the proportion of students passing with a C- or greater increased at a declining rate, although the effects were still positive across all quartiles. These patterns hold for both English and math courses. My results suggest that many students who would have been placed in remedial education would still be able to pass college-level courses at similar rates to students who would have not been placed in remedial education, even those at higher levels of predicted probability of requiring remedial education. These negligible effects of allowing direct access to college-level courses can help guide other community colleges with respect to managing the costly, and seemingly ineffective, remedial education system.